# Cloud Security Breaches: every page as markdown A sourced ledger of public cloud incidents from 2019 on, read by access path. Company statements, filings and post-mortems only. --- # Cloud security breaches, read by the access path that was used Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. The Cloud Breach Ledger records public cloud incidents from 2019 to 16 November 2026 and reads each one by the access path used: leaked keys, stolen credentials, over-permissioned identities, public resources, cross-tenant flaws, third-party access, exposed services or retained access. Every fact comes from a primary source. The Cloud Breach Ledger: 111 incidents. - Leaked access keys and secrets: 27 - Stolen user credentials and session tokens: 47 - Over-permissioned identities: 13 - Public storage, databases and services: 15 - Cross-tenant access: 3 - Third-party access: 25 - Exposed network services and request paths: 8 - Retained access after departure or change: 7 An entry can count under two access paths, and 11 entries have no path stated, so the counts above do not add up to 111. Every entry rests on a company statement, a regulator filing, a court document or a provider post-mortem; a field no such source states reads "not public". Workload vulnerabilities and posture scanning are outside the ledger: it records the access path an incident used. Corrections go to editor@cloudsecuritybreaches.com; each entry has its own correction link. ## Sources - Veradigm Inc.: Veradigm Form 8-K, Item 8.01, filed 2026-09-08: https://www.sec.gov/Archives/edgar/data/1124804/000119312526385249/mdrx-20260908.htm (read 16 Nov 2026) - IEH Corporation: IEH Corporation Form 8-K, Item 8.01: https://www.sec.gov/Archives/edgar/data/50292/000117494726000761/form8k-36187_iehc.htm (read 16 Nov 2026) - Amgen: Amgen Inc. Form 8-K, Item 1.05, July 2026: https://www.sec.gov/Archives/edgar/data/318154/000031815426000119/amgn-20260729.htm (read 16 Nov 2026) - AdaptHealth Corp.: AdaptHealth Form 8-K, Item 1.05, filed 2026-07-02: https://www.sec.gov/Archives/edgar/data/1725255/000110465926080297/ahco-20260627x8k.htm (read 16 Nov 2026) - AdaptHealth: AdaptHealth Corp. Form 8-K, Item 1.05, June 2026: https://www.sec.gov/Archives/edgar/data/1725255/000110465926080297/ahco-20260627x8k.htm (read 16 Nov 2026) - 8x8, Inc.: 8x8, Inc. Form 8-K, Item 1.05: https://www.sec.gov/Archives/edgar/data/1023731/000102373126000084/eght-20260617.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/ --- # The Cloud Breach Ledger: public cloud incidents by access path, edition 16 Nov 2026 Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. The Cloud Breach Ledger lists public cloud incidents from 2019 to 16 November 2026. Each entry states what was exposed, how access was gained and which of eight access paths applies, using only company statements, regulator filings, court documents and provider post-mortems. A field without such a source reads not public. The sourcing rule has no exceptions. An entry's facts come from a company statement (a press release, blog post, customer notice or SEC filing by the company), a regulator filing or decision, a court document, or a provider post-mortem. News articles, vendor research blogs and encyclopedias are not sources, though they may lead to one. Each entry quotes one verbatim sentence from its primary source, and an incident that cannot be sourced this way is left out. Not public means that no primary source states the fact as of the read date, 16 November 2026. It does not mean the fact is unknown to the company or hidden from anyone. When a company or regulator publishes the fact later, the field changes and the change is logged on the corrections page. The ledger names organizations only, writes no motive or blame, and uses a group name only where the company's own statement uses it. Counts are computed from the entries on the date of the edition. A count is the number of entries that match a filter, such as one access path, one provider or one year of publication. Counts measure sourced incidents, not all incidents, and a higher count for a provider does not say it is less safe. Entries are listed by date and never ranked. Third-party figures are kept apart, quoted with their publisher and year. ## Sources an entry may rest on | Kind | Entries citing it | | --- | --- | | company statement | 39 | | regulator filing | 60 | | court document | 6 | | provider post mortem | 14 | ## The eight access paths | Path | Definition | | --- | --- | | [Leaked access keys and secrets](/access-paths/leaked-access-key/) | A leaked access key is a long-lived key, token or secret that ended up somewhere it should not be, such as a code repository, a container image, a log or a public file, and was then used by someone other than its owner to call a cloud provider's interfaces or reach a service. | | [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) | Stolen user credentials are the password or session token of a person, obtained by phishing, an infostealer, password spraying or a similar method, and then used to sign in as that person to a cloud console, a software-as-a-service tenant or a data platform, often where no second factor was required. | | [Over-permissioned identities](/access-paths/over-permissioned-identity/) | An over-permissioned identity is a user, role, service account or application that held more access than its job needed, so that the extra access, and not the way the identity was compromised, set how much data or infrastructure an incident reached and how long the intruder could keep moving. | | [Public storage, databases and services](/access-paths/public-resource/) | A public resource is a storage container, database or service that was left reachable to anyone who could find its address, with no sign-in required, because a setting, a policy or a firewall rule allowed anonymous access, whether by choice, by mistake or through a later change. | | [Cross-tenant access](/access-paths/cross-tenant/) | Cross-tenant access is a flaw in a provider's service, or a setting in a customer's application or account, that let one customer's tenant reach resources belonging to another customer, so the boundary the provider draws around each customer did not hold for that request or token. | | [Third-party access](/access-paths/third-party-access/) | Third-party access is access to a customer's cloud or software-as-a-service data that came through a vendor, supplier, integration or contractor the customer had trusted, so the incident began in the outside party's systems or credentials and reached the customer through a grant already in place. | | [Exposed network services and request paths](/access-paths/exposed-network-service/) | An exposed network service is a reachable service or request path, such as an open management port, an unpatched internet-facing application or a request forged to reach a metadata endpoint, that gave a caller a way into a workload and into the credentials or data attached to it. | | [Retained access after departure or change](/access-paths/retained-access/) | Retained access is access that a former employee, contractor or application kept after the reason for it ended, such as a password, key or connected app that nobody removed, and which was later used or left available to be used against the organization's cloud accounts. | Cite this line: 'Cloud Security Breaches, the Cloud Breach Ledger, edition 16 Nov 2026, https://cloudsecuritybreaches.com/methodology/' ## Related pages - [Cloud security statistics: ledger counts and quoted figures](/statistics/) - [Correct an entry in the Cloud Breach Ledger](/ledger-corrections/) - [30 questions about public cloud incidents](/faq/) ## Sources - AWS IAM User Guide: manage access keys for IAM users: https://docs.aws.amazon.com/IAM/latest/UserGuide/id_credentials_access-keys.html (read 16 Nov 2026) - AWS IAM User Guide: temporary security credentials: https://docs.aws.amazon.com/IAM/latest/UserGuide/id_credentials_temp.html (read 16 Nov 2026) - Microsoft Learn: shared access signatures overview: https://learn.microsoft.com/en-us/azure/storage/common/storage-sas-overview (read 16 Nov 2026) - Google Cloud: best practices for managing service account keys: https://cloud.google.com/iam/docs/best-practices-for-managing-service-account-keys (read 16 Nov 2026) - Microsoft Learn: how multifactor authentication works: https://learn.microsoft.com/en-us/entra/identity/authentication/concept-mfa-howitworks (read 16 Nov 2026) - AWS IAM User Guide: multi-factor authentication in IAM: https://docs.aws.amazon.com/IAM/latest/UserGuide/id_credentials_mfa.html (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/methodology/ --- # Veradigm Inc. (2026): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Veradigm Inc. disclosed the incident on 8 Sep 2026. What was exposed, as the sources state it: The filing says copies of certain personal data of patients were downloaded, including in some instances Social Security numbers; no clinical or medical data was involved and a small number of customers were affected. - **Disclosed:** 8 Sep 2026 - **Incident window:** not public - **Cloud or provider:** not public - **What was exposed:** The filing says copies of certain personal data of patients were downloaded, including in some instances Social Security numbers; no clinical or medical data was involved and a small number of customers were affected. - **How access was gained:** The filing says a third-party vendor had an incident and an unauthorized party obtained credentials from the vendor’s environment to a company application programming interface, then used them to download patient data; the compromised credentials gave access only through that interface. - **Access path:** [Third-party access](/access-paths/third-party-access/); also [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** issue vendors narrowly scoped, short-lived API credentials, restrict them by network origin and alert on bulk downloads through vendor interfaces ## In the words of the source “an unauthorized party obtained credentials from the vendor’s environment to a Company application programming interface used by the vendor to provide services on behalf of the Company’s customers.” (regulator filing, [Veradigm Form 8-K, Item 8.01, filed 2026-09-08](https://www.sec.gov/Archives/edgar/data/1124804/000119312526385249/mdrx-20260908.htm)) ## Primary sources - regulator filing: [Veradigm Form 8-K, Item 8.01, filed 2026-09-08](https://www.sec.gov/Archives/edgar/data/1124804/000119312526385249/mdrx-20260908.htm), dated 8 Sep 2026 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Veradigm%20Inc.%20(veradigm-vendor-api-2026)) with the entry name in the subject. ## Related pages - [IEH Corporation 2026: microsoft 365](/incidents/ieh-microsoft-365-2026/) - [Amgen 2026: cloud environments](/incidents/amgen-cloud-environments-2026/) - [AdaptHealth Corp. 2026: cloud apps](/incidents/adapthealth-cloud-apps-2026/) ## Sources - Veradigm Form 8-K, Item 8.01, filed 2026-09-08: https://www.sec.gov/Archives/edgar/data/1124804/000119312526385249/mdrx-20260908.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/veradigm-vendor-api-2026/ --- # IEH Corporation (2026): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. IEH Corporation disclosed the incident on 6 Aug 2026. What was exposed, as the sources state it: Mailbox contents of one employee, including email messages, attachments, customer communications, purchase orders and engineering-related documentation; the filing says no evidence currently exists that data was exfiltrated. - **Disclosed:** 6 Aug 2026 - **Incident window:** August 4, 2026 - **Cloud or provider:** Microsoft 365 - **What was exposed:** Mailbox contents of one employee, including email messages, attachments, customer communications, purchase orders and engineering-related documentation; the filing says no evidence currently exists that data was exfiltrated - **How access was gained:** The filing says a phishing email carried a link disguised as a Microsoft document-sharing link, and the user entered Microsoft 365 credentials into a fraudulent login page. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** require phishing-resistant multi-factor authentication for every Microsoft 365 account and block sign-in from unmanaged devices ## In the words of the source “The user accessed the link and entered Microsoft 365 credentials into a fraudulent login page” (regulator filing, [IEH Corporation Form 8-K, Item 8.01](https://www.sec.gov/Archives/edgar/data/50292/000117494726000761/form8k-36187_iehc.htm)) ## Primary sources - regulator filing: [IEH Corporation Form 8-K, Item 8.01](https://www.sec.gov/Archives/edgar/data/50292/000117494726000761/form8k-36187_iehc.htm), dated 6 Aug 2026 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20IEH%20Corporation%20(ieh-microsoft-365-2026)) with the entry name in the subject. ## Related pages - [Amgen 2026: cloud environments](/incidents/amgen-cloud-environments-2026/) - [AdaptHealth Corp. 2026: cloud apps](/incidents/adapthealth-cloud-apps-2026/) - [AdaptHealth 2026: contractor session](/incidents/adapthealth-contractor-session-2026/) ## Sources - IEH Corporation Form 8-K, Item 8.01: https://www.sec.gov/Archives/edgar/data/50292/000117494726000761/form8k-36187_iehc.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/ieh-microsoft-365-2026/ --- # Amgen (2026): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Amgen disclosed the incident on 31 Jul 2026. What was exposed, as the sources state it: The company stated some of its data, including proprietary data, patient protected health information and other information, was exfiltrated from the cloud environments. Counts are not public. - **Disclosed:** 31 Jul 2026 - **Incident window:** July 2026 (the filing) - **Cloud or provider:** not public - **What was exposed:** The company stated some of its data, including proprietary data, patient protected health information and other information, was exfiltrated from the cloud environments. Counts are not public. - **How access was gained:** The Form 8-K says the company identified unauthorized activity involving data stored in cloud environments hosted by third-party cloud service providers. How access was gained is not public. - **Access path:** Not public: the sources do not state how access was gained - **Kind of control that addresses it:** not public (the filing does not state how access was gained) ## In the words of the source “In July 2026, Amgen Inc. (the Company) identified unauthorized activity involving data stored in cloud environments hosted by third-party cloud service providers.” (regulator filing, [Amgen Inc. Form 8-K, Item 1.05, July 2026](https://www.sec.gov/Archives/edgar/data/318154/000031815426000119/amgn-20260729.htm)) ## Primary sources - regulator filing: [Amgen Inc. Form 8-K, Item 1.05, July 2026](https://www.sec.gov/Archives/edgar/data/318154/000031815426000119/amgn-20260729.htm), dated 31 Jul 2026 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Amgen%20(amgen-cloud-environments-2026)) with the entry name in the subject. ## Related pages - [AdaptHealth Corp. 2026: cloud apps](/incidents/adapthealth-cloud-apps-2026/) - [AdaptHealth 2026: contractor session](/incidents/adapthealth-contractor-session-2026/) - [8x8, Inc. 2026: salesforce klue](/incidents/8x8-salesforce-klue-2026/) ## Sources - Amgen Inc. Form 8-K, Item 1.05, July 2026: https://www.sec.gov/Archives/edgar/data/318154/000031815426000119/amgn-20260729.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/amgen-cloud-environments-2026/ --- # AdaptHealth Corp. (2026): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. AdaptHealth Corp. disclosed the incident on 2 Jul 2026. What was exposed, as the sources state it: The filing says a stored password file associated with insurance billing was exfiltrated, plus passwords associated with insurance billing and certain personal and protected health information of patients; the full scope was not yet determined. - **Disclosed:** 2 Jul 2026 - **Incident window:** data claim received June 15, 2026; materiality determined June 27, 2026 - **Cloud or provider:** not public - **What was exposed:** The filing says a stored password file associated with insurance billing was exfiltrated, plus passwords associated with insurance billing and certain personal and protected health information of patients; the full scope was not yet determined. - **How access was gained:** The filing says a threat actor gained unauthorized access to certain cloud-based business applications (the provider is not named) after a social engineering attack that compromised a user session associated with a third-party contractor. - **Access path:** [Third-party access](/access-paths/third-party-access/); also [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** require phishing-resistant multi-factor authentication and short session lifetimes for contractor accounts, and revoke sessions on any report of social engineering ## In the words of the source “The incident was the result of a successful social engineering attack that compromised a user session associated with a third-party contractor.” (regulator filing, [AdaptHealth Form 8-K, Item 1.05, filed 2026-07-02](https://www.sec.gov/Archives/edgar/data/1725255/000110465926080297/ahco-20260627x8k.htm)) ## Primary sources - regulator filing: [AdaptHealth Form 8-K, Item 1.05, filed 2026-07-02](https://www.sec.gov/Archives/edgar/data/1725255/000110465926080297/ahco-20260627x8k.htm), dated 2 Jul 2026 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20AdaptHealth%20Corp.%20(adapthealth-cloud-apps-2026)) with the entry name in the subject. ## Related pages - [AdaptHealth 2026: contractor session](/incidents/adapthealth-contractor-session-2026/) - [8x8, Inc. 2026: salesforce klue](/incidents/8x8-salesforce-klue-2026/) - [iRhythm Holdings, Inc. 2026: hosted apps](/incidents/irhythm-hosted-apps-2026/) ## Sources - AdaptHealth Form 8-K, Item 1.05, filed 2026-07-02: https://www.sec.gov/Archives/edgar/data/1725255/000110465926080297/ahco-20260627x8k.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/adapthealth-cloud-apps-2026/ --- # AdaptHealth (2026): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. AdaptHealth disclosed the incident on 2 Jul 2026. What was exposed, as the sources state it: The company stated a stored password file associated with insurance billing was exfiltrated, external electronic health record portals were accessed, and affected data includes passwords associated with insurance billing and certain personal and protected health information of patients. - **Disclosed:** 2 Jul 2026 - **Incident window:** Statement of June 15, 2026 received from the actor; materiality determined June 27, 2026 - **Cloud or provider:** not public - **What was exposed:** The company stated a stored password file associated with insurance billing was exfiltrated, external electronic health record portals were accessed, and affected data includes passwords associated with insurance billing and certain personal and protected health information of patients. - **How access was gained:** The Form 8-K says a threat actor gained unauthorized access to certain cloud-based business applications, and that the incident was the result of a successful social engineering attack that compromised a user session associated with a third-party contractor. - **Access path:** [Third-party access](/access-paths/third-party-access/); also [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** require phishing-resistant authentication and short-lived sessions for contractor accounts, and limit contractor access to the systems they need ## In the words of the source “The incident was the result of a successful social engineering attack that compromised a user session associated with a third-party contractor.” (regulator filing, [AdaptHealth Corp. Form 8-K, Item 1.05, June 2026](https://www.sec.gov/Archives/edgar/data/1725255/000110465926080297/ahco-20260627x8k.htm)) ## Primary sources - regulator filing: [AdaptHealth Corp. Form 8-K, Item 1.05, June 2026](https://www.sec.gov/Archives/edgar/data/1725255/000110465926080297/ahco-20260627x8k.htm), dated 2 Jul 2026 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20AdaptHealth%20(adapthealth-contractor-session-2026)) with the entry name in the subject. ## Related pages - [8x8, Inc. 2026: salesforce klue](/incidents/8x8-salesforce-klue-2026/) - [iRhythm Holdings, Inc. 2026: hosted apps](/incidents/irhythm-hosted-apps-2026/) - [SailPoint, Inc. 2026: github](/incidents/sailpoint-github-2026/) ## Sources - AdaptHealth Corp. Form 8-K, Item 1.05, June 2026: https://www.sec.gov/Archives/edgar/data/1725255/000110465926080297/ahco-20260627x8k.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/adapthealth-contractor-session-2026/ --- # 8x8, Inc. (2026): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. 8x8, Inc. disclosed the incident on 23 Jun 2026. What was exposed, as the sources state it: Certain information from the company's Salesforce system: fragmented contract and opportunity information, sales team notes, and contact information of customers and prospective customers. - **Disclosed:** 23 Jun 2026 - **Incident window:** June 11 to 12, 2026 - **Cloud or provider:** Salesforce - **What was exposed:** Certain information from the company's Salesforce system: fragmented contract and opportunity information, sales team notes, and contact information of customers and prospective customers - **How access was gained:** The filing says an unauthorized third party threat actor exploited the Klue third-party application programming integration connected to the company's Salesforce customer relationship management system. - **Access path:** [Third-party access](/access-paths/third-party-access/) - **Kind of control that addresses it:** inventory and least-privilege every third-party integration connected to the CRM, and alert on integration activity outside its normal pattern ## In the words of the source “The threat actor gained unauthorized access and exfiltrated certain information from 8x8’s Salesforce system.” (regulator filing, [8x8, Inc. Form 8-K, Item 1.05](https://www.sec.gov/Archives/edgar/data/1023731/000102373126000084/eght-20260617.htm)) ## Primary sources - regulator filing: [8x8, Inc. Form 8-K, Item 1.05](https://www.sec.gov/Archives/edgar/data/1023731/000102373126000084/eght-20260617.htm), dated 23 Jun 2026 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%208x8%2C%20Inc.%20(8x8-salesforce-klue-2026)) with the entry name in the subject. ## Related pages - [iRhythm Holdings, Inc. 2026: hosted apps](/incidents/irhythm-hosted-apps-2026/) - [SailPoint, Inc. 2026: github](/incidents/sailpoint-github-2026/) - [A Maryland medical system (called Company A in the indictm](/incidents/maryland-medical-system-m365-2016/) ## Sources - 8x8, Inc. Form 8-K, Item 1.05: https://www.sec.gov/Archives/edgar/data/1023731/000102373126000084/eght-20260617.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/8x8-salesforce-klue-2026/ --- # iRhythm Holdings, Inc. (2026): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. iRhythm Holdings, Inc. disclosed the incident on 15 Jun 2026. What was exposed, as the sources state it: The filing says a threat actor claimed sensitive information including proprietary data, patient protected health information and other personal information, and the company confirmed certain data was exfiltrated; the volume is not public. - **Disclosed:** 15 Jun 2026 - **Incident window:** identified June 8, 2026, as the filing says - **Cloud or provider:** not public - **What was exposed:** The filing says a threat actor claimed sensitive information including proprietary data, patient protected health information and other personal information, and the company confirmed certain data was exfiltrated; the volume is not public. - **How access was gained:** The filing says the affected data was obtained through social engineering from certain third-party-hosted business applications (the provider is not named). - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** require phishing-resistant multi-factor authentication for every account on hosted business applications and verify help-desk callers through a second channel ## In the words of the source “the affected data was obtained through social engineering and is from certain third-party-hosted business applications.” (regulator filing, [iRhythm Form 8-K, Item 1.05, filed 2026-06-15](https://www.sec.gov/Archives/edgar/data/1388658/000138865826000055/irtc-20260610.htm)) ## Primary sources - regulator filing: [iRhythm Form 8-K, Item 1.05, filed 2026-06-15](https://www.sec.gov/Archives/edgar/data/1388658/000138865826000055/irtc-20260610.htm), dated 15 Jun 2026 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20iRhythm%20Holdings%2C%20Inc.%20(irhythm-hosted-apps-2026)) with the entry name in the subject. ## Related pages - [SailPoint, Inc. 2026: github](/incidents/sailpoint-github-2026/) - [A Maryland medical system (called Company A in the indictm](/incidents/maryland-medical-system-m365-2016/) - [ADT Inc. 2026: cloud environments](/incidents/adt-cloud-environments-2026/) ## Sources - iRhythm Form 8-K, Item 1.05, filed 2026-06-15: https://www.sec.gov/Archives/edgar/data/1388658/000138865826000055/irtc-20260610.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/irhythm-hosted-apps-2026/ --- # SailPoint, Inc. (2026): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. SailPoint, Inc. disclosed the incident on 8 May 2026. What was exposed, as the sources state it: A subset of the company’s GitHub repositories; the filing says no evidence that customer data in production or staging environments was accessed, and customers with information in the accessed repositories were notified. - **Disclosed:** 8 May 2026 - **Incident window:** detected April 20, 2026, as the filing says - **Cloud or provider:** GitHub - **What was exposed:** A subset of the company’s GitHub repositories; the filing says no evidence that customer data in production or staging environments was accessed, and customers with information in the accessed repositories were notified. - **How access was gained:** The filing says the company detected unauthorized access to a subset of its GitHub repositories and that the root cause was a vulnerability in a third-party application, which it does not name. - **Access path:** [Third-party access](/access-paths/third-party-access/) - **Kind of control that addresses it:** review the scopes of every third-party application connected to source-control organizations, remove unused integrations, and patch or retire vulnerable ones ## In the words of the source “The root cause was a vulnerability in a third-party application, which has been remediated.” (regulator filing, [SailPoint Form 8-K, Item 7.01, filed 2026-05-08](https://www.sec.gov/Archives/edgar/data/2030781/000162828026033016/sail-20260508.htm)) ## Primary sources - regulator filing: [SailPoint Form 8-K, Item 7.01, filed 2026-05-08](https://www.sec.gov/Archives/edgar/data/2030781/000162828026033016/sail-20260508.htm), dated 8 May 2026 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20SailPoint%2C%20Inc.%20(sailpoint-github-2026)) with the entry name in the subject. ## Related pages - [A Maryland medical system (called Company A in the indictm](/incidents/maryland-medical-system-m365-2016/) - [ADT Inc. 2026: cloud environments](/incidents/adt-cloud-environments-2026/) - [People Central Pte. Ltd. 2026: pdpc](/incidents/people-central-pdpc-2024/) ## Sources - SailPoint Form 8-K, Item 7.01, filed 2026-05-08: https://www.sec.gov/Archives/edgar/data/2030781/000162828026033016/sail-20260508.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/sailpoint-github-2026/ --- # A Maryland medical system (called Company A in the indictment release) (2026): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. A Maryland medical system (called Company A in the indictment release) disclosed the incident on 1 May 2026. What was exposed, as the sources state it: The release says usernames, passwords, cookies, images, videos and other data of 195 victims, who were current or former employees or people affiliated with them, were obtained. - **Disclosed:** 1 May 2026 - **Incident window:** July 2016 to September 2024, as the release says - **Cloud or provider:** Microsoft 365 - **What was exposed:** The release says usernames, passwords, cookies, images, videos and other data of 195 victims, who were current or former employees or people affiliated with them, were obtained. - **How access was gained:** The release says an employee used keylogging, cookie managers and mailbox rules to reach victims’ personal and professional accounts, including Microsoft 365, and exported browser cookies to keep unauthorized access from other devices. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/); also [Retained access after departure or change](/access-paths/retained-access/) - **Kind of control that addresses it:** alert on mailbox rules that delete security notices, bind sessions to managed devices and monitor privileged staff with access to endpoint data ## In the words of the source “used various cyber intrusion techniques — such as keylogging, cookie managers, mailbox-rule creation, and file masquerading — to obtain access to personal and professional accounts of people who were current or former employees” (court document, [U.S. Attorney District of Maryland: Maryland Man Indicted on Unauthorized Computer Access Related to a Maryland Medical System and Identity Theft Charges](https://web.archive.org/web/2026/https://www.justice.gov/usao-md/pr/maryland-man-indicted-unauthorized-computer-access-related-maryland-medical-system-and)) ## Primary sources - court document: [U.S. Attorney District of Maryland: Maryland Man Indicted on Unauthorized Computer Access Related to a Maryland Medical System and Identity Theft Charges](https://web.archive.org/web/2026/https://www.justice.gov/usao-md/pr/maryland-man-indicted-unauthorized-computer-access-related-maryland-medical-system-and), dated 1 May 2026 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20A%20Maryland%20medical%20system%20(called%20Company%20A%20in%20the%20indictment%20release)%20(maryland-medical-system-m365-2016)) with the entry name in the subject. ## Related pages - [ADT Inc. 2026: cloud environments](/incidents/adt-cloud-environments-2026/) - [People Central Pte. Ltd. 2026: pdpc](/incidents/people-central-pdpc-2024/) - [Mixpanel 2025: smishing](/incidents/mixpanel-smishing-2025/) ## Sources - U.S. Attorney District of Maryland: Maryland Man Indicted on Unauthorized Computer Access Related to a Maryland Medical System and Identity Theft Charges: https://web.archive.org/web/2026/https://www.justice.gov/usao-md/pr/maryland-man-indicted-unauthorized-computer-access-related-maryland-medical-system-and (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/maryland-medical-system-m365-2016/ --- # ADT Inc. (2026): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. ADT Inc. disclosed the incident on 24 Apr 2026. What was exposed, as the sources state it: The filing says only limited customer and prospective customer data was accessed. The cloud provider is not named. The sources do not state how access was gained, so the access path reads not public. - **Disclosed:** 24 Apr 2026 - **Incident window:** Became aware on April 20, 2026 - **Cloud or provider:** not public - **What was exposed:** The filing says only limited customer and prospective customer data was accessed. The cloud provider is not named. - **How access was gained:** The filing says the company became aware of unauthorized access to certain cloud-based environments, terminated the access, and notified law enforcement. How access was gained is not public. - **Access path:** Not public: the sources do not state how access was gained - **Kind of control that addresses it:** not public (the sources do not state how access was gained) ## In the words of the source “On April 20, 2026, ADT Inc. (ADT or the Company) became aware of unauthorized access to certain cloud-based environments.” (regulator filing, [ADT Inc. Form 8-K, Item 8.01, filed 2026-04-24](https://www.sec.gov/Archives/edgar/data/1703056/000170305626000038/adt-20260420.htm)) ## Primary sources - regulator filing: [ADT Inc. Form 8-K, Item 8.01, filed 2026-04-24](https://www.sec.gov/Archives/edgar/data/1703056/000170305626000038/adt-20260420.htm), dated 24 Apr 2026 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20ADT%20Inc.%20(adt-cloud-environments-2026)) with the entry name in the subject. ## Related pages - [People Central Pte. Ltd. 2026: pdpc](/incidents/people-central-pdpc-2024/) - [Mixpanel 2025: smishing](/incidents/mixpanel-smishing-2025/) - [Illuminate Education, Inc. 2025: retained access after dep](/incidents/illuminate-education-2021/) ## Sources - ADT Inc. Form 8-K, Item 8.01, filed 2026-04-24: https://www.sec.gov/Archives/edgar/data/1703056/000170305626000038/adt-20260420.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/adt-cloud-environments-2026/ --- # People Central Pte. Ltd. (2026): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. People Central Pte. Ltd. disclosed the incident on 8 Jan 2026. The ledger reads the access path as exposed network services and request paths. The cloud or provider involved: AWS. Every field rests on the primary sources listed below. Fields no primary source states read not public. - **Disclosed:** 8 Jan 2026 - **Incident window:** 29 April 2024 (extortion email); discovered by 3 May 2024 - **Cloud or provider:** AWS - **What was exposed:** The PDPC decision states personal data of 95,000 employees of the company's clients was put at risk (including name, identity card number, salary, bank account number and address), plus 24,765 emergency contacts and children. The threat actor deleted databases and likely exfiltrated data, and data was found for sale on the dark web. - **How access was gained:** The decision says the company, a cloud-based HR software provider, had Remote Desktop Protocol access open to the internet without two-factor authentication on its AWS cloud servers, settings allowed inbound traffic from all IP addresses in the cloud environment, and SQL injection vulnerabilities were present with no web application firewall. - **Access path:** [Exposed network services and request paths](/access-paths/exposed-network-service/) - **Kind of control that addresses it:** close management ports to the internet, restrict cloud network rules to known addresses, require MFA for remote access, and test web applications for injection ## In the words of the source “Remote Desktop Protocol ( RDP ) access had been open to the internet and did not require two-factor authentication ( 2FA ).” (regulator filing, [PDPC decision: People Central Pte. Ltd.](https://www.pdpc.gov.sg/assets/f9fdd00e-2e3d-4e69-a017-4da0db04d1ad)) ## Primary sources - regulator filing: [PDPC decision: People Central Pte. Ltd.](https://www.pdpc.gov.sg/assets/f9fdd00e-2e3d-4e69-a017-4da0db04d1ad), dated 8 Jan 2026 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20People%20Central%20Pte.%20Ltd.%20(people-central-pdpc-2024)) with the entry name in the subject. ## Related pages - [Mixpanel 2025: smishing](/incidents/mixpanel-smishing-2025/) - [Illuminate Education, Inc. 2025: retained access after dep](/incidents/illuminate-education-2021/) - [GitHub (npm registry) 2025: shai hulud](/incidents/github-npm-shai-hulud-2025/) ## Sources - PDPC decision: People Central Pte. Ltd.: https://www.pdpc.gov.sg/assets/f9fdd00e-2e3d-4e69-a017-4da0db04d1ad (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/people-central-pdpc-2024/ --- # Mixpanel (2025): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Mixpanel disclosed the incident on 27 Nov 2025. What was exposed, as the sources state it: not public. The ledger reads the access path as stolen user credentials and session tokens. The cloud or provider involved: Mixpanel. Every field rests on the primary sources listed below. - **Disclosed:** 27 Nov 2025 - **Incident window:** from November 8, 2025 - **Cloud or provider:** Mixpanel - **What was exposed:** not public - **How access was gained:** The company stated it detected a smishing campaign on November 8, 2025, and took steps to contain unauthorized access and secure impacted user accounts; it revoked sessions and rotated compromised credentials for impacted accounts. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** require phishing-resistant multi-factor authentication for staff accounts and revoke sessions on any report of credential phishing ## In the words of the source “On November 8th, 2025, Mixpanel detected a smishing campaign and promptly executed our incident response processes.” (company statement, [Our response to a recent security incident](https://mixpanel.com/blog/sms-security-incident/)) ## Primary sources - company statement: [Our response to a recent security incident](https://mixpanel.com/blog/sms-security-incident/), dated 27 Nov 2025 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Mixpanel%20(mixpanel-smishing-2025)) with the entry name in the subject. ## Related pages - [Illuminate Education, Inc. 2025: retained access after dep](/incidents/illuminate-education-2021/) - [GitHub (npm registry) 2025: shai hulud](/incidents/github-npm-shai-hulud-2025/) - [Cloudflare 2025: salesforce drift](/incidents/cloudflare-salesforce-drift-2025/) ## Sources - Our response to a recent security incident: https://mixpanel.com/blog/sms-security-incident/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/mixpanel-smishing-2025/ --- # Illuminate Education, Inc. (2025): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Illuminate Education, Inc. disclosed the incident on 6 Nov 2025. The ledger reads the access path as retained access after departure or change with stolen user credentials and session tokens. The cloud or provider involved: not public (the FTC says a third-party cloud provider, not named in the press release). - **Disclosed:** 6 Nov 2025 - **Incident window:** Late December 2021 (the FTC press release); the New York Attorney General release dates the breach to 2022 - **Cloud or provider:** not public (the FTC says a third-party cloud provider, not named in the press release) - **What was exposed:** The FTC states the hacker gained access to personal data of 10.1 million students, including email and mailing addresses, dates of birth, student records and health-related information. The New York Attorney General states the breach exposed personal information of millions of students, 1.7 million of them in New York. - **How access was gained:** The FTC press release says that, according to the complaint, a hacker used the credentials of a former employee, who had left the company three and a half years earlier, to breach databases stored on a third-party cloud provider. It also says a third-party vendor had alerted the company in January 2020 to numerous security vulnerabilities and that student data was stored in plain text until at least January 2022. - **Access path:** [Retained access after departure or change](/access-paths/retained-access/); also [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** disable every account and credential of a departing employee at the moment of departure, and review active identities for dormant accounts on a schedule ## In the words of the source “In its complaint, the FTC alleged that in late December 2021 a hacker used the credentials of a former employee—who had departed Illuminate three and a half years prior—to breach Illuminate’s databases stored on a third-party cloud provider” (regulator filing, [FTC press release: FTC Takes Action Against Education Technology Provider for Failing to Secure Students’ Personal Data (web.archive.org copy)](https://web.archive.org/web/2024/https://www.ftc.gov/news-events/news/press-releases/2025/12/ftc-takes-action-against-education-technology-provider-failing-secure-students-personal-data)) ## Primary sources - regulator filing: [FTC press release: FTC Takes Action Against Education Technology Provider for Failing to Secure Students’ Personal Data (web.archive.org copy)](https://web.archive.org/web/2024/https://www.ftc.gov/news-events/news/press-releases/2025/12/ftc-takes-action-against-education-technology-provider-failing-secure-students-personal-data), dated 1 Dec 2025 - regulator filing: [New York Attorney General press release: Attorney General James and Multistate Coalition Secure $5.1 Million from Education Software Company for Failing to Protect Students’ Data](https://ag.ny.gov/press-release/2025/attorney-general-james-and-multistate-coalition-secure-51-million-education), dated 6 Nov 2025 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Illuminate%20Education%2C%20Inc.%20(illuminate-education-2021)) with the entry name in the subject. ## Related pages - [GitHub (npm registry) 2025: shai hulud](/incidents/github-npm-shai-hulud-2025/) - [Cloudflare 2025: salesforce drift](/incidents/cloudflare-salesforce-drift-2025/) - [Nx (Nrwl) 2025: github actions](/incidents/nx-github-actions-2025/) ## Sources - FTC press release: FTC Takes Action Against Education Technology Provider for Failing to Secure Students’ Personal Data (web.archive.org copy): https://web.archive.org/web/2024/https://www.ftc.gov/news-events/news/press-releases/2025/12/ftc-takes-action-against-education-technology-provider-failing-secure-students-personal-data (read 16 Nov 2026) - New York Attorney General press release: Attorney General James and Multistate Coalition Secure $5.1 Million from Education Software Company for Failing to Protect Students’ Data: https://ag.ny.gov/press-release/2025/attorney-general-james-and-multistate-coalition-secure-51-million-education (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/illuminate-education-2021/ --- # GitHub (npm registry) (2025): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. GitHub (npm registry) disclosed the incident on 22 Sep 2025. What was exposed, as the sources state it: The post says 500+ compromised packages were removed from the npm registry and that the worm could steal multiple types of secrets, not just npm tokens. - **Disclosed:** 22 Sep 2025 - **Incident window:** notified September 14, 2025, as the post says - **Cloud or provider:** GitHub - **What was exposed:** The post says 500+ compromised packages were removed from the npm registry and that the worm could steal multiple types of secrets, not just npm tokens. - **How access was gained:** GitHub stated that a self-replicating worm spread through compromised maintainer accounts by injecting malicious post-install scripts into popular packages. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/); also [Leaked access keys and secrets](/access-paths/leaked-access-key/) - **Kind of control that addresses it:** require phishing-resistant multi-factor authentication and short-lived scoped tokens for package publishing and disable install scripts by default ## In the words of the source “a self-replicating worm that infiltrated the npm ecosystem via compromised maintainer accounts by injecting malicious post-install scripts into popular JavaScript packages.” (provider post mortem, [Our plan for a more secure npm supply chain](https://github.blog/security/supply-chain-security/our-plan-for-a-more-secure-npm-supply-chain/)) ## Primary sources - provider post mortem: [Our plan for a more secure npm supply chain](https://github.blog/security/supply-chain-security/our-plan-for-a-more-secure-npm-supply-chain/), dated 22 Sep 2025 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20GitHub%20(npm%20registry)%20(github-npm-shai-hulud-2025)) with the entry name in the subject. ## Related pages - [Cloudflare 2025: salesforce drift](/incidents/cloudflare-salesforce-drift-2025/) - [Nx (Nrwl) 2025: github actions](/incidents/nx-github-actions-2025/) - [Salesloft (Drift) 2025: leaked access keys and secrets](/incidents/salesloft-drift-2025/) ## Sources - Our plan for a more secure npm supply chain: https://github.blog/security/supply-chain-security/our-plan-for-a-more-secure-npm-supply-chain/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/github-npm-shai-hulud-2025/ --- # Cloudflare (2025): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Cloudflare disclosed the incident on 2 Sep 2025. What was exposed, as the sources state it: Salesforce case objects (customer support tickets and associated data: contact information, subject lines and case correspondence, not attachments); the company found 104 Cloudflare API tokens in the data. - **Disclosed:** 2 Sep 2025 - **Incident window:** August 9 to 17, 2025 - **Cloud or provider:** Salesforce - **What was exposed:** Salesforce case objects (customer support tickets and associated data: contact information, subject lines and case correspondence, not attachments); the company found 104 Cloudflare API tokens in the data - **How access was gained:** The company stated the threat actor obtained OAuth credentials associated with the Salesloft Drift chat agent's Salesforce integration, and used a stolen credential used by the Salesloft integration to log in to its Salesforce tenant. - **Access path:** [Third-party access](/access-paths/third-party-access/); also [Leaked access keys and secrets](/access-paths/leaked-access-key/) - **Kind of control that addresses it:** restrict third-party integration tokens to approved source addresses, give them minimum scopes, and rotate them on a short schedule ## In the words of the source “a threat actor breached their systems. As part of the breach, the threat actor was able to obtain OAuth credentials associated with the Salesloft Drift chat agent’s Salesforce integration” (company statement, [The impact of the Salesloft Drift breach on Cloudflare and our customers](https://blog.cloudflare.com/response-to-salesloft-drift-incident/)) ## Primary sources - company statement: [The impact of the Salesloft Drift breach on Cloudflare and our customers](https://blog.cloudflare.com/response-to-salesloft-drift-incident/), dated 2 Sep 2025 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Cloudflare%20(cloudflare-salesforce-drift-2025)) with the entry name in the subject. ## Related pages - [Nx (Nrwl) 2025: github actions](/incidents/nx-github-actions-2025/) - [Salesloft (Drift) 2025: leaked access keys and secrets](/incidents/salesloft-drift-2025/) - [Google 2025: salesforce instance](/incidents/google-salesforce-instance-2025/) ## Sources - The impact of the Salesloft Drift breach on Cloudflare and our customers: https://blog.cloudflare.com/response-to-salesloft-drift-incident/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/cloudflare-salesforce-drift-2025/ --- # Nx (Nrwl) (2025): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Nx (Nrwl) disclosed the incident on 26 Aug 2025. What was exposed, as the sources state it: The company's npm publishing token; malicious package versions scanned users' systems and uploaded results to public GitHub repositories. The company stated Nx Cloud was not affected. - **Disclosed:** 26 Aug 2025 - **Incident window:** August 26, 2025 (malicious packages live for 4 hours) - **Cloud or provider:** GitHub - **What was exposed:** The company's npm publishing token; malicious package versions scanned users' systems and uploaded results to public GitHub repositories. The company stated Nx Cloud was not affected. - **How access was gained:** The post-mortem says attackers exploited a GitHub Actions injection vulnerability to steal the npm publishing token; the repository's workflow permissions were read and write. - **Access path:** [Leaked access keys and secrets](/access-paths/leaked-access-key/); also [Over-permissioned identities](/access-paths/over-permissioned-identity/) - **Kind of control that addresses it:** set CI workflow tokens to read-only by default, avoid running untrusted pull request input with repository secrets, and publish with short-lived trusted-publisher credentials ## In the words of the source “Attackers exploited a GitHub Actions injection vulnerability to steal our NPM publishing token and publish malicious packages for 4 hours.” (provider post mortem, [S1ngularity - What Happened, How We Responded, What We Learned](https://nx.dev/blog/s1ngularity-postmortem)) ## Primary sources - provider post mortem: [S1ngularity - What Happened, How We Responded, What We Learned](https://nx.dev/blog/s1ngularity-postmortem), dated 5 Sep 2025 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Nx%20(Nrwl)%20(nx-github-actions-2025)) with the entry name in the subject. ## Related pages - [Salesloft (Drift) 2025: leaked access keys and secrets](/incidents/salesloft-drift-2025/) - [Google 2025: salesforce instance](/incidents/google-salesforce-instance-2025/) - [Amazon Web Services 2025: aws q extension](/incidents/aws-amazon-q-extension-2025/) ## Sources - S1ngularity - What Happened, How We Responded, What We Learned: https://nx.dev/blog/s1ngularity-postmortem (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/nx-github-actions-2025/ --- # Salesloft (Drift) (2025): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Salesloft (Drift) disclosed the incident on 20 Aug 2025. What was exposed, as the sources state it: OAuth tokens for Drift customers' technology integrations, obtained from the Drift AWS environment; Salesloft GitHub repositories downloaded; environment variable secrets exfiltrated. The ledger reads the access path as leaked access keys and secrets with third-party access. - **Disclosed:** 20 Aug 2025 - **Incident window:** March 22 to September 5, 2025 - **Cloud or provider:** AWS - **What was exposed:** OAuth tokens for Drift customers' technology integrations, obtained from the Drift AWS environment; Salesloft GitHub repositories downloaded; environment variable secrets exfiltrated - **How access was gained:** The investigation summary says the threat actor accessed the Salesloft GitHub account (March to June 2025), then accessed Drift's AWS environment and obtained OAuth tokens for customers' integrations, and used them to access data via Drift integrations. - **Access path:** [Leaked access keys and secrets](/access-paths/leaked-access-key/); also [Third-party access](/access-paths/third-party-access/) - **Kind of control that addresses it:** scope and monitor source-code hosting tokens, keep integration secrets out of repository and build environments, and restrict where integration tokens may be used from ## In the words of the source “The threat actor then accessed Drift’s AWS environment and obtained OAuth tokens for Drift customers’ technology integrations.” (company statement, [Salesloft trust site: Drift security update and Mandiant investigation summary](https://trust.salesloft.com/?uid=Drift+Security+Update)) ## Primary sources - company statement: [Salesloft trust site: Drift security update and Mandiant investigation summary](https://trust.salesloft.com/?uid=Drift+Security+Update), dated not stated If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Salesloft%20(Drift)%20(salesloft-drift-2025)) with the entry name in the subject. ## Related pages - [Google 2025: salesforce instance](/incidents/google-salesforce-instance-2025/) - [Amazon Web Services 2025: aws q extension](/incidents/aws-amazon-q-extension-2025/) - [Zoomcar Holdings, Inc. 2025: cloud](/incidents/zoomcar-cloud-2025/) ## Sources - Salesloft trust site: Drift security update and Mandiant investigation summary: https://trust.salesloft.com/?uid=Drift+Security+Update (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/salesloft-drift-2025/ --- # Google (2025): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Google disclosed the incident on 5 Aug 2025. What was exposed, as the sources state it: The update says a corporate Salesforce instance used to store contact information and related notes for small and medium businesses was affected; data retrieved was confined to basic and largely publicly available business information such as business names and contact details. - **Disclosed:** 5 Aug 2025 - **Incident window:** June 2025, as the update says - **Cloud or provider:** Salesforce - **What was exposed:** The update says a corporate Salesforce instance used to store contact information and related notes for small and medium businesses was affected; data retrieved was confined to basic and largely publicly available business information such as business names and contact details. - **How access was gained:** Google stated that the instance was impacted by UNC6040 activity described in the same post, which describes voice phishing in which employees are tricked into authorizing a malicious connected app or sharing credentials. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/); also [Third-party access](/access-paths/third-party-access/) - **Kind of control that addresses it:** restrict who may authorize connected apps, require phishing-resistant multi-factor authentication and limit logins to known network ranges ## In the words of the source “In June, one of Google’s corporate Salesforce instances was impacted by similar UNC6040 activity described in this post.” (company statement, [The Cost of a Call: From Voice Phishing to Data Extortion (Update, August 5)](https://cloud.google.com/blog/topics/threat-intelligence/voice-phishing-data-extortion)) ## Primary sources - company statement: [The Cost of a Call: From Voice Phishing to Data Extortion (Update, August 5)](https://cloud.google.com/blog/topics/threat-intelligence/voice-phishing-data-extortion), dated 5 Aug 2025 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Google%20(google-salesforce-instance-2025)) with the entry name in the subject. ## Related pages - [Amazon Web Services 2025: aws q extension](/incidents/aws-amazon-q-extension-2025/) - [Zoomcar Holdings, Inc. 2025: cloud](/incidents/zoomcar-cloud-2025/) - [The Walt Disney Company 2025: slack](/incidents/disney-slack-2024/) ## Sources - The Cost of a Call: From Voice Phishing to Data Extortion (Update, August 5): https://cloud.google.com/blog/topics/threat-intelligence/voice-phishing-data-extortion (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/google-salesforce-instance-2025/ --- # Amazon Web Services (2025): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Amazon Web Services disclosed the incident on 23 Jul 2025. What was exposed, as the sources state it: Malicious code was committed to the open-source repository of the Amazon Q Developer for VS Code extension and distributed in version 1.84.0; AWS stated it was unsuccessful in executing and made no changes to services or customer environments. - **Disclosed:** 23 Jul 2025 - **Incident window:** not public - **Cloud or provider:** AWS - **What was exposed:** Malicious code was committed to the open-source repository of the Amazon Q Developer for VS Code extension and distributed in version 1.84.0; AWS stated it was unsuccessful in executing and made no changes to services or customer environments - **How access was gained:** The bulletin says the extension had an inappropriately scoped GitHub token in its CodeBuild configuration, and with that token the threat actor committed malicious code. - **Access path:** [Over-permissioned identities](/access-paths/over-permissioned-identity/); also [Leaked access keys and secrets](/access-paths/leaked-access-key/) - **Kind of control that addresses it:** give build-system tokens the narrowest repository permissions and require review before any commit can enter a release ## In the words of the source “With that access token, the threat actor was able to commit malicious code into the extension's open-source repository that was automatically included in a release.” (provider post mortem, [Security Update for Amazon Q Developer Extension for Visual Studio Code (Version #1.84), AWS-2025-015](https://aws.amazon.com/security/security-bulletins/AWS-2025-015/)) ## Primary sources - provider post mortem: [Security Update for Amazon Q Developer Extension for Visual Studio Code (Version #1.84), AWS-2025-015](https://aws.amazon.com/security/security-bulletins/AWS-2025-015/), dated 23 Jul 2025 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Amazon%20Web%20Services%20(aws-amazon-q-extension-2025)) with the entry name in the subject. ## Related pages - [Zoomcar Holdings, Inc. 2025: cloud](/incidents/zoomcar-cloud-2025/) - [The Walt Disney Company 2025: slack](/incidents/disney-slack-2024/) - [Nova Scotia Power 2025: opc](/incidents/nova-scotia-power-opc-2025/) ## Sources - Security Update for Amazon Q Developer Extension for Visual Studio Code (Version #1.84), AWS-2025-015: https://aws.amazon.com/security/security-bulletins/AWS-2025-015/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/aws-amazon-q-extension-2025/ --- # Zoomcar Holdings, Inc. (2025): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Zoomcar Holdings, Inc. disclosed the incident on 13 Jun 2025. The sources do not state how access was gained, so the access path reads not public. The cloud or provider involved: not public. Every field rests on the primary sources listed below. - **Disclosed:** 13 Jun 2025 - **Incident window:** Identified on June 9, 2025 - **Cloud or provider:** not public - **What was exposed:** The filing says an unauthorized third party accessed a limited dataset with personal information of a subset of approximately 8.4 million users (names, phone numbers, car registration numbers, personal addresses and email addresses). It says there was no evidence that financial information or plaintext passwords were compromised. - **How access was gained:** The filing says the company identified unauthorized access to its information systems after employees received communications from a threat actor, and lists additional safeguards across the cloud and internal network as a response. How access was gained and which cloud provider was used are not public. - **Access path:** Not public: the sources do not state how access was gained - **Kind of control that addresses it:** not public (the sources do not state how access was gained) ## In the words of the source “the Company determined that an unauthorized third party accessed a limited dataset containing certain personal information of a subset of approximately 8.4 million users” (regulator filing, [Zoomcar Holdings Form 8-K, Item 1.05, filed 2025-06-13](https://www.sec.gov/Archives/edgar/data/1854275/000121390025054319/ea0245724-8k_zoomcar.htm)) ## Primary sources - regulator filing: [Zoomcar Holdings Form 8-K, Item 1.05, filed 2025-06-13](https://www.sec.gov/Archives/edgar/data/1854275/000121390025054319/ea0245724-8k_zoomcar.htm), dated 13 Jun 2025 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Zoomcar%20Holdings%2C%20Inc.%20(zoomcar-cloud-2025)) with the entry name in the subject. ## Related pages - [The Walt Disney Company 2025: slack](/incidents/disney-slack-2024/) - [Nova Scotia Power 2025: opc](/incidents/nova-scotia-power-opc-2025/) - [Advanced Computer Software Group Ltd (health and care subs](/incidents/advanced-computer-software-2022/) ## Sources - Zoomcar Holdings Form 8-K, Item 1.05, filed 2025-06-13: https://www.sec.gov/Archives/edgar/data/1854275/000121390025054319/ea0245724-8k_zoomcar.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/zoomcar-cloud-2025/ --- # The Walt Disney Company (2025): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. The Walt Disney Company disclosed the incident on 1 May 2025. What was exposed, as the sources state it: The release says approximately 1.1 terabytes of confidential data from thousands of Slack channels was downloaded, then published in July 2024. - **Disclosed:** 1 May 2025 - **Incident window:** April to May 2024, as the release says - **Cloud or provider:** Slack - **What was exposed:** The release says approximately 1.1 terabytes of confidential data from thousands of Slack channels was downloaded, then published in July 2024. - **How access was gained:** The release says a malicious file posted online let an individual into an employee’s personal computer, where stored login credentials were obtained and used to reach the employee’s Slack account. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** require phishing-resistant multi-factor authentication and device checks for chat and collaboration accounts and expire sessions on unmanaged devices ## In the words of the source “obtaining login information, and using that information to illegally download confidential data from the Burbank-based mass media and entertainment conglomerate via the employee’s Slack online communications account.” (court document, [U.S. Attorney CDCA: Santa Clarita Man Agrees to Plead Guilty to Hacking Disney Employee’s Computer, Downloading Confidential Data from Company](https://web.archive.org/web/2026/https://www.justice.gov/usao-cdca/pr/santa-clarita-man-agrees-plead-guilty-hacking-disney-employees-computer-downloading)) ## Primary sources - court document: [U.S. Attorney CDCA: Santa Clarita Man Agrees to Plead Guilty to Hacking Disney Employee’s Computer, Downloading Confidential Data from Company](https://web.archive.org/web/2026/https://www.justice.gov/usao-cdca/pr/santa-clarita-man-agrees-plead-guilty-hacking-disney-employees-computer-downloading), dated 1 May 2025 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20The%20Walt%20Disney%20Company%20(disney-slack-2024)) with the entry name in the subject. ## Related pages - [Nova Scotia Power 2025: opc](/incidents/nova-scotia-power-opc-2025/) - [Advanced Computer Software Group Ltd (health and care subs](/incidents/advanced-computer-software-2022/) - [People Central Pte. Ltd. (Singapore, cloud-based HR softwa](/incidents/people-central-aws-rdp-2024/) ## Sources - U.S. Attorney CDCA: Santa Clarita Man Agrees to Plead Guilty to Hacking Disney Employee’s Computer, Downloading Confidential Data from Company: https://web.archive.org/web/2026/https://www.justice.gov/usao-cdca/pr/santa-clarita-man-agrees-plead-guilty-hacking-disney-employees-computer-downloading (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/disney-slack-2024/ --- # Nova Scotia Power (2025): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Nova Scotia Power disclosed the incident on 28 Apr 2025. The ledger reads the access path as stolen user credentials and session tokens with over-permissioned identities. The cloud or provider involved: not public. Every field rests on the primary sources listed below. - **Disclosed:** 28 Apr 2025 - **Incident window:** 19 March to 25 April 2025 - **Cloud or provider:** not public - **What was exposed:** The OPC compliance letter states about 375,000 current and 540,000 former customers were affected: names, phone numbers, emails, mailing addresses, dates of birth, account histories (including billing and bank account numbers), driver's license numbers and social insurance numbers. It says data was exfiltrated from on-premises network files and cloud storage. - **How access was gained:** The letter says an employee visited a compromised website and clicked a link, malware was installed, the threat actor moved across systems using accounts with domain administrator privileges, harvested credentials, took data between 23 and 25 April 2025, then used harvested credentials to destroy backups and deploy ransomware. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/); also [Over-permissioned identities](/access-paths/over-permissioned-identity/) - **Kind of control that addresses it:** limit and monitor domain administrator accounts, segment cloud storage from the on-premises network, keep protected backups, and delete personal data that is no longer needed ## In the words of the source “Between April 23 and April 25, 2025, the threat actor exfiltrated data from on-premises network files and cloud storage.” (regulator filing, [OPC Compliance Letter by Nova Scotia Power](https://www.priv.gc.ca/en/opc-actions-and-decisions/investigations/investigations-into-businesses/2026/2026-ns-power-ca/)) ## Primary sources - regulator filing: [OPC Compliance Letter by Nova Scotia Power](https://www.priv.gc.ca/en/opc-actions-and-decisions/investigations/investigations-into-businesses/2026/2026-ns-power-ca/), dated not stated If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Nova%20Scotia%20Power%20(nova-scotia-power-opc-2025)) with the entry name in the subject. ## Related pages - [Advanced Computer Software Group Ltd (health and care subs](/incidents/advanced-computer-software-2022/) - [People Central Pte. Ltd. (Singapore, cloud-based HR softwa](/incidents/people-central-aws-rdp-2024/) - [Salesforce customers (multiple organizations) 2025: vishin](/incidents/salesforce-customers-vishing-2025/) ## Sources - OPC Compliance Letter by Nova Scotia Power: https://www.priv.gc.ca/en/opc-actions-and-decisions/investigations/investigations-into-businesses/2026/2026-ns-power-ca/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/nova-scotia-power-opc-2025/ --- # Advanced Computer Software Group Ltd (health and care subsidiary) (2025): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Advanced Computer Software Group Ltd (health and care subsidiary) disclosed the incident on 27 Mar 2025. What was exposed, as the sources state it: The ICO states personal information belonging to 79,404 people was taken, including details of how to gain entry into the homes of 890 people receiving care at home. - **Disclosed:** 27 Mar 2025 - **Incident window:** August 2022 - **Cloud or provider:** not public - **What was exposed:** The ICO states personal information belonging to 79,404 people was taken, including details of how to gain entry into the homes of 890 people receiving care at home. - **How access was gained:** The ICO states hackers accessed certain systems of the company’s health and care subsidiary via a customer account that did not have multi-factor authentication (MFA). The ICO’s investigation found gaps in the deployment of MFA, a lack of comprehensive vulnerability scanning and inadequate patch management. The ICO does not state how the account was obtained. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** require multi-factor authentication on every customer and administrator account that can reach a hosted service, with no exceptions, and verify coverage continuously ## In the words of the source “Hackers accessed certain systems of Advanced’s health and care subsidiary via a customer account that did not have multi-factor authentication (MFA).” (regulator filing, [ICO news: Software provider fined £3m following 2022 ransomware attack](https://ico.org.uk/about-the-ico/media-centre/news-and-blogs/2025/03/software-provider-fined-3m-following-2022-ransomware-attack/)) ## Primary sources - regulator filing: [ICO news: Software provider fined £3m following 2022 ransomware attack](https://ico.org.uk/about-the-ico/media-centre/news-and-blogs/2025/03/software-provider-fined-3m-following-2022-ransomware-attack/), dated 27 Mar 2025 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Advanced%20Computer%20Software%20Group%20Ltd%20(health%20and%20care%20subsidiary)%20(advanced-computer-software-2022)) with the entry name in the subject. ## Related pages - [People Central Pte. Ltd. (Singapore, cloud-based HR softwa](/incidents/people-central-aws-rdp-2024/) - [Salesforce customers (multiple organizations) 2025: vishin](/incidents/salesforce-customers-vishing-2025/) - [PowerSchool 2025: powersource](/incidents/powerschool-powersource-2024/) ## Sources - ICO news: Software provider fined £3m following 2022 ransomware attack: https://ico.org.uk/about-the-ico/media-centre/news-and-blogs/2025/03/software-provider-fined-3m-following-2022-ransomware-attack/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/advanced-computer-software-2022/ --- # People Central Pte. Ltd. (Singapore, cloud-based HR software provider) (2025): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. People Central Pte. Ltd. (Singapore, cloud-based HR software provider) disclosed the incident on 20 Mar 2025. The ledger reads the access path as exposed network services and request paths. The cloud or provider involved: Amazon Web Services. Every field rests on the primary sources listed below. - **Disclosed:** 20 Mar 2025 - **Incident window:** April 29, 2024 extortion email; regulator notified May 3, 2024 - **Cloud or provider:** Amazon Web Services - **What was exposed:** The decision says a threat actor deleted databases and likely exfiltrated data from the company's AWS servers; data of 95,000 employees of its clients (including NRIC number, salary and bank account number) and of 24,765 emergency contacts and children was put at risk, and some data was found for sale on the dark web. - **How access was gained:** The decision lists contributing lapses: SQL injection vulnerabilities in the web application with no web application firewall, RDP access open to the internet without two-factor authentication, settings that allowed inbound traffic from all IP addresses in the cloud environment, and vulnerability scanning only every two years. - **Access path:** [Exposed network services and request paths](/access-paths/exposed-network-service/) - **Kind of control that addresses it:** do not expose remote-administration ports to the internet, require multi-factor authentication for any administrative access to cloud servers, and restrict security groups to known addresses ## In the words of the source “access had been open to the internet and did not require two-factor authentication” (regulator filing, [Singapore Personal Data Protection Commission decision: People Central Pte Ltd, [2025] SGPDPCS 4, 20 March 2025 (date from the document title)](https://www.pdpc.gov.sg/assets/f9fdd00e-2e3d-4e69-a017-4da0db04d1ad?v=1)) ## Primary sources - regulator filing: [Singapore Personal Data Protection Commission decision: People Central Pte Ltd, [2025] SGPDPCS 4, 20 March 2025 (date from the document title)](https://www.pdpc.gov.sg/assets/f9fdd00e-2e3d-4e69-a017-4da0db04d1ad?v=1), dated 20 Mar 2025 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20People%20Central%20Pte.%20Ltd.%20(Singapore%2C%20cloud-based%20HR%20software%20provider)%20(people-central-aws-rdp-2024)) with the entry name in the subject. ## Related pages - [Salesforce customers (multiple organizations) 2025: vishin](/incidents/salesforce-customers-vishing-2025/) - [PowerSchool 2025: powersource](/incidents/powerschool-powersource-2024/) - [BeyondTrust 2024: remote support saas](/incidents/beyondtrust-remote-support-saas-2024/) ## Sources - Singapore Personal Data Protection Commission decision: People Central Pte Ltd, [2025] SGPDPCS 4, 20 March 2025 (date from the document title): https://www.pdpc.gov.sg/assets/f9fdd00e-2e3d-4e69-a017-4da0db04d1ad?v=1 (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/people-central-aws-rdp-2024/ --- # Salesforce customers (multiple organizations) (2025): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Salesforce customers (multiple organizations) disclosed the incident on 12 Mar 2025. What was exposed, as the sources state it: Data exfiltrated from customer Salesforce accounts through a connected app; counts not public. The ledger reads the access path as stolen user credentials and session tokens with third-party access. - **Disclosed:** 12 Mar 2025 - **Incident window:** not public - **Cloud or provider:** Salesforce - **What was exposed:** Data exfiltrated from customer Salesforce accounts through a connected app; counts not public - **How access was gained:** Salesforce stated threat actors impersonated IT support by phone, luring customers' employees and third-party support workers to phishing pages to steal credentials and MFA tokens, or to add a malicious connected app. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/); also [Third-party access](/access-paths/third-party-access/) - **Kind of control that addresses it:** restrict logins to known network ranges, require phishing-resistant MFA, and limit who may authorize connected apps ## In the words of the source “Once the threat actor gains access to a customer’s Salesforce account or adds a connected app, they use the connected app to exfiltrate data.” (provider post mortem, [Protect Your Salesforce Environment from Social Engineering Threats](https://www.salesforce.com/blog/protect-against-social-engineering/)) ## Primary sources - provider post mortem: [Protect Your Salesforce Environment from Social Engineering Threats](https://www.salesforce.com/blog/protect-against-social-engineering/), dated 12 Mar 2025 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Salesforce%20customers%20(multiple%20organizations)%20(salesforce-customers-vishing-2025)) with the entry name in the subject. ## Related pages - [PowerSchool 2025: powersource](/incidents/powerschool-powersource-2024/) - [BeyondTrust 2024: remote support saas](/incidents/beyondtrust-remote-support-saas-2024/) - [Avaya Holdings Corp. 2024: cloud file sharing](/incidents/avaya-cloud-file-sharing-2020/) ## Sources - Protect Your Salesforce Environment from Social Engineering Threats: https://www.salesforce.com/blog/protect-against-social-engineering/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/salesforce-customers-vishing-2025/ --- # PowerSchool (2025): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. PowerSchool disclosed the incident on 7 Jan 2025. What was exposed, as the sources state it: The page says exfiltration of select PowerSchool SIS instances of Students and Teachers tables; counts not public in this source. The ledger reads the access path as stolen user credentials and session tokens. - **Disclosed:** 7 Jan 2025 - **Incident window:** became aware December 28, 2024, as the page says - **Cloud or provider:** PowerSchool - **What was exposed:** The page says exfiltration of select PowerSchool SIS instances of Students and Teachers tables; counts not public in this source. - **How access was gained:** The company stated, citing its forensic firm’s final report, that the threat actor accessed PowerSource, a community-focused customer support portal, using a single compromised credential. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** require phishing-resistant multi-factor authentication on every support and maintenance portal that can reach customer tenants, and limit what such portals can export ## In the words of the source “The Threat Actor accessed PowerSource, a community-focused customer support portal, using a single compromised credential.” (company statement, [PowerSchool Cybersecurity Incident](https://www.powerschool.com/security/sis-incident/)) ## Primary sources - company statement: [PowerSchool Cybersecurity Incident](https://www.powerschool.com/security/sis-incident/), dated 7 Jan 2025 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20PowerSchool%20(powerschool-powersource-2024)) with the entry name in the subject. ## Related pages - [BeyondTrust 2024: remote support saas](/incidents/beyondtrust-remote-support-saas-2024/) - [Avaya Holdings Corp. 2024: cloud file sharing](/incidents/avaya-cloud-file-sharing-2020/) - [Verkada Inc. 2024: over-permissioned identities](/incidents/verkada-2021/) ## Sources - PowerSchool Cybersecurity Incident: https://www.powerschool.com/security/sis-incident/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/powerschool-powersource-2024/ --- # BeyondTrust (2024): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. BeyondTrust disclosed the incident on 8 Dec 2024. The ledger reads the access path as exposed network services and request paths with leaked access keys and secrets. The cloud or provider involved: Amazon Web Services. Every field rests on the primary sources listed below. - **Disclosed:** 8 Dec 2024 - **Incident window:** Identified December 5, 2024; investigation completed January 17, 2025 - **Cloud or provider:** Amazon Web Services - **What was exposed:** The company stated a security incident involved 17 Remote Support SaaS customers: an infrastructure API key was used to enable access to certain Remote Support SaaS instances by resetting local application passwords. It stated no other BeyondTrust products or systems were affected and ransomware was not involved. - **How access was gained:** The company stated a zero-day vulnerability of a third-party application was used to gain access to an online asset in a BeyondTrust AWS account, which allowed the threat actor to obtain an infrastructure API key that was then used against a separate AWS account that operated Remote Support infrastructure. - **Access path:** [Exposed network services and request paths](/access-paths/exposed-network-service/); also [Leaked access keys and secrets](/access-paths/leaked-access-key/) - **Kind of control that addresses it:** patch internet-facing third-party applications quickly, keep infrastructure API keys out of reach of application hosts, and scope keys so one account's key cannot act in a separate production account ## In the words of the source “The investigation determined that a zero-day vulnerability of a third-party application was used to gain access to an online asset in a BeyondTrust AWS account.” (company statement, [BeyondTrust Remote Support SaaS Service Security Investigation (updated 2025-01-28)](https://www.beyondtrust.com/remote-support-saas-service-security-investigation)) ## Primary sources - company statement: [BeyondTrust Remote Support SaaS Service Security Investigation (updated 2025-01-28)](https://www.beyondtrust.com/remote-support-saas-service-security-investigation), dated 8 Dec 2024 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20BeyondTrust%20(beyondtrust-remote-support-saas-2024)) with the entry name in the subject. ## Related pages - [Avaya Holdings Corp. 2024: cloud file sharing](/incidents/avaya-cloud-file-sharing-2020/) - [Verkada Inc. 2024: over-permissioned identities](/incidents/verkada-2021/) - [AT&T 2024: att third party cloud](/incidents/att-third-party-cloud-2024/) ## Sources - BeyondTrust Remote Support SaaS Service Security Investigation (updated 2025-01-28): https://www.beyondtrust.com/remote-support-saas-service-security-investigation (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/beyondtrust-remote-support-saas-2024/ --- # Avaya Holdings Corp. (2024): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Avaya Holdings Corp. disclosed the incident on 22 Oct 2024. What was exposed, as the sources state it: The SEC order, as described in the SEC press release, finds the threat actor accessed at least 145 files in the company’s cloud file sharing environment, in addition to a limited number of company email messages. - **Disclosed:** 22 Oct 2024 - **Incident window:** The SEC says the company learned of the access in 2020 - **Cloud or provider:** not public - **What was exposed:** The SEC order, as described in the SEC press release, finds the threat actor accessed at least 145 files in the company’s cloud file sharing environment, in addition to a limited number of company email messages. - **How access was gained:** The SEC press release says the threat actor likely behind the SolarWinds Orion hack accessed the company’s systems without authorization. It describes the access as part of an investigation of public companies potentially impacted by the compromise of SolarWinds’ Orion software. The SEC charged the company over the accuracy of its public disclosure of the incident. - **Access path:** [Third-party access](/access-paths/third-party-access/) - **Kind of control that addresses it:** treat vendor-supplied management software as a privileged path: restrict its network reach and credentials, and monitor the cloud file-sharing environment for bulk access by identities that do not normally read it ## In the words of the source “the threat actor likely behind the SolarWinds Orion hack had accessed their systems without authorization” (regulator filing, [SEC press release 2024-174: SEC Charges Four Companies With Misleading Cyber Disclosures](https://www.sec.gov/newsroom/press-releases/2024-174)) ## Primary sources - regulator filing: [SEC press release 2024-174: SEC Charges Four Companies With Misleading Cyber Disclosures](https://www.sec.gov/newsroom/press-releases/2024-174), dated 22 Oct 2024 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Avaya%20Holdings%20Corp.%20(avaya-cloud-file-sharing-2020)) with the entry name in the subject. ## Related pages - [Verkada Inc. 2024: over-permissioned identities](/incidents/verkada-2021/) - [AT&T 2024: att third party cloud](/incidents/att-third-party-cloud-2024/) - [Advance Auto Parts 2024: not public](/incidents/advance-auto-parts-2024/) ## Sources - SEC press release 2024-174: SEC Charges Four Companies With Misleading Cyber Disclosures: https://www.sec.gov/newsroom/press-releases/2024-174 (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/avaya-cloud-file-sharing-2020/ --- # Verkada Inc. (2024): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Verkada Inc. disclosed the incident on 30 Aug 2024. The ledger reads the access path as over-permissioned identities with exposed network services and request paths. The cloud or provider involved: Verkada (cloud-based security camera provider). Every field rests on the primary sources listed below. - **Disclosed:** 30 Aug 2024 - **Incident window:** Between December 2020 and March 2021 (at least two security breaches, per the FTC) - **Cloud or provider:** Verkada (cloud-based security camera provider) - **What was exposed:** The FTC states that in the March 2021 breach the hacker had access to over 150,000 live Verkada customer cameras and to customer information such as physical addresses, audio recordings and customer WiFi credentials. The Justice Department states security-camera footage of consumers at locations such as hospitals and schools was exposed to unauthorized access. - **How access was gained:** The FTC press release says the complaint alleges the company failed to provide appropriate security measures, including failing to require unique and complex passwords, adequately encrypt customer data and implement secure network controls. The Justice Department release says the complaint alleges failures in appropriate access management and data protection controls. Neither source states how the hacker obtained the access. - **Access path:** [Over-permissioned identities](/access-paths/over-permissioned-identity/); also [Exposed network services and request paths](/access-paths/exposed-network-service/) - **Kind of control that addresses it:** enforce unique, complex credentials and least-privilege access management on the administrative identities of a cloud service, and segment internal systems from its customer-facing platform ## In the words of the source “For example, the company failed to require unique and complex passwords, adequately encrypt customer data, and implement secure network controls.” (regulator filing, [FTC press release: FTC Takes Action Against Security Camera Firm Verkada over Charges it Failed to Secure Videos, Other Personal Data and Violated CAN-SPAM Act (web.archive.org copy)](https://web.archive.org/web/2025/https://www.ftc.gov/news-events/news/press-releases/2024/08/ftc-takes-action-against-security-camera-firm-verkada-over-charges-it-failed-secure-videos-other)) ## Primary sources - regulator filing: [FTC press release: FTC Takes Action Against Security Camera Firm Verkada over Charges it Failed to Secure Videos, Other Personal Data and Violated CAN-SPAM Act (web.archive.org copy)](https://web.archive.org/web/2025/https://www.ftc.gov/news-events/news/press-releases/2024/08/ftc-takes-action-against-security-camera-firm-verkada-over-charges-it-failed-secure-videos-other), dated 30 Aug 2024 - court document: [Justice Department press release: $2.95M Penalty and Permanent Injunction Resolves Lawsuit Against Verkada Inc. (complaint in the U.S. District Court for the Northern District of California; web.archive.org copy)](https://web.archive.org/web/2024/https://www.justice.gov/archives/opa/pr/295m-penalty-and-permanent-injunction-resolves-lawsuit-against-verkada-inc-alleged-unlawful), dated 4 Sep 2024 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Verkada%20Inc.%20(verkada-2021)) with the entry name in the subject. ## Related pages - [AT&T 2024: att third party cloud](/incidents/att-third-party-cloud-2024/) - [Advance Auto Parts 2024: not public](/incidents/advance-auto-parts-2024/) - [Pure Storage 2024: snowflake workspace](/incidents/pure-storage-snowflake-workspace-2024/) ## Sources - FTC press release: FTC Takes Action Against Security Camera Firm Verkada over Charges it Failed to Secure Videos, Other Personal Data and Violated CAN-SPAM Act (web.archive.org copy): https://web.archive.org/web/2025/https://www.ftc.gov/news-events/news/press-releases/2024/08/ftc-takes-action-against-security-camera-firm-verkada-over-charges-it-failed-secure-videos-other (read 16 Nov 2026) - Justice Department press release: $2.95M Penalty and Permanent Injunction Resolves Lawsuit Against Verkada Inc. (complaint in the U.S. District Court for the Northern District of California; web.archive.org copy): https://web.archive.org/web/2024/https://www.justice.gov/archives/opa/pr/295m-penalty-and-permanent-injunction-resolves-lawsuit-against-verkada-inc-alleged-unlawful (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/verkada-2021/ --- # AT&T (2024): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. AT&T disclosed the incident on 12 Jul 2024. What was exposed, as the sources state it: The filing says files containing AT&T records of customer call and text interactions from approximately May 1 to October 31, 2022, and January 2, 2023, were exfiltrated. - **Disclosed:** 12 Jul 2024 - **Incident window:** April 14 to April 25, 2024, as the filing says - **Cloud or provider:** not public - **What was exposed:** The filing says files containing AT&T records of customer call and text interactions from approximately May 1 to October 31, 2022, and January 2, 2023, were exfiltrated. - **How access was gained:** The filing says threat actors unlawfully accessed an AT&T workspace on a third-party cloud platform; the filing does not say how access was gained. - **Access path:** Not public: the sources do not state how access was gained - **Kind of control that addresses it:** not public (the sources do not state how access was gained) ## In the words of the source “Based on its investigation, AT&T believes that threat actors unlawfully accessed an AT&T workspace on a third-party cloud platform” (regulator filing, [AT&T Form 8-K, Item 1.05, filed 2024-07-12](https://www.sec.gov/Archives/edgar/data/732717/000073271724000046/t-20240506.htm)) ## Primary sources - regulator filing: [AT&T Form 8-K, Item 1.05, filed 2024-07-12](https://www.sec.gov/Archives/edgar/data/732717/000073271724000046/t-20240506.htm), dated 12 Jul 2024 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20AT%26T%20(att-third-party-cloud-2024)) with the entry name in the subject. ## Related pages - [Advance Auto Parts 2024: not public](/incidents/advance-auto-parts-2024/) - [Pure Storage 2024: snowflake workspace](/incidents/pure-storage-snowflake-workspace-2024/) - [Live Nation Entertainment 2024: ticketmaster](/incidents/live-nation-ticketmaster-2024/) ## Sources - AT&T Form 8-K, Item 1.05, filed 2024-07-12: https://www.sec.gov/Archives/edgar/data/732717/000073271724000046/t-20240506.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/att-third-party-cloud-2024/ --- # Advance Auto Parts (2024): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Advance Auto Parts disclosed the incident on 14 Jun 2024. What was exposed, as the sources state it: The filing says a criminal threat actor offered what it alleged to be Company data for sale on June 4, 2024; the filing does not state what data was exposed. - **Disclosed:** 14 Jun 2024 - **Incident window:** Identified May 23, 2024, as the filing says; start not public - **Cloud or provider:** not public - **What was exposed:** The filing says a criminal threat actor offered what it alleged to be Company data for sale on June 4, 2024; the filing does not state what data was exposed. - **How access was gained:** The filing says the Company identified unauthorized activity within a third-party cloud database environment containing Company data; the filing does not say how access was gained. - **Access path:** Not public: the sources do not state how access was gained - **Kind of control that addresses it:** not public (the sources do not state how access was gained) ## In the words of the source “On May 23, 2024, Advance Auto Parts, Inc. (the Company) identified unauthorized activity within a third-party cloud database environment containing Company data” (regulator filing, [Advance Auto Parts Form 8-K, Item 8.01, filed 2024-06-14](https://www.sec.gov/Archives/edgar/data/1158449/000115844924000162/aap-20240523.htm)) ## Primary sources - regulator filing: [Advance Auto Parts Form 8-K, Item 8.01, filed 2024-06-14](https://www.sec.gov/Archives/edgar/data/1158449/000115844924000162/aap-20240523.htm), dated 14 Jun 2024 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Advance%20Auto%20Parts%20(advance-auto-parts-2024)) with the entry name in the subject. ## Related pages - [Pure Storage 2024: snowflake workspace](/incidents/pure-storage-snowflake-workspace-2024/) - [Live Nation Entertainment 2024: ticketmaster](/incidents/live-nation-ticketmaster-2024/) - [Hugging Face 2024: huggingface spaces secrets](/incidents/huggingface-spaces-secrets-2024/) ## Sources - Advance Auto Parts Form 8-K, Item 8.01, filed 2024-06-14: https://www.sec.gov/Archives/edgar/data/1158449/000115844924000162/aap-20240523.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/advance-auto-parts-2024/ --- # Pure Storage (2024): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Pure Storage disclosed the incident on 13 Jun 2024. What was exposed, as the sources state it: The company stated the workspace did not include compromising information such as passwords for customer systems or any of the data stored on customer systems, but contained telemetry information used for proactive customer support. - **Disclosed:** 13 Jun 2024 - **Incident window:** June 2024 (the filing) - **Cloud or provider:** Snowflake - **What was exposed:** The company stated the workspace did not include compromising information such as passwords for customer systems or any of the data stored on customer systems, but contained telemetry information used for proactive customer support. - **How access was gained:** The filing says a third party temporarily gained unauthorized access to a single Snowflake data analytics workspace. How the access was gained is not public in the filing. - **Access path:** Not public: the sources do not state how access was gained - **Kind of control that addresses it:** not public (the sources do not state how access was gained) ## In the words of the source “in June 2024, we were the subject of a security incident involving a third party that temporarily gained unauthorized access to a single Snowflake data analytics workspace.” (regulator filing, [Pure Storage, Inc. Form 10-Q for the quarter ended May 5, 2024](https://www.sec.gov/Archives/edgar/data/1474432/000162828024027955/pstg-20240505.htm)) ## Primary sources - regulator filing: [Pure Storage, Inc. Form 10-Q for the quarter ended May 5, 2024](https://www.sec.gov/Archives/edgar/data/1474432/000162828024027955/pstg-20240505.htm), dated 13 Jun 2024 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Pure%20Storage%20(pure-storage-snowflake-workspace-2024)) with the entry name in the subject. ## Related pages - [Live Nation Entertainment 2024: ticketmaster](/incidents/live-nation-ticketmaster-2024/) - [Hugging Face 2024: huggingface spaces secrets](/incidents/huggingface-spaces-secrets-2024/) - [Snowflake customer accounts (multiple organizations) 2024:](/incidents/snowflake-customer-accounts-2024/) ## Sources - Pure Storage, Inc. Form 10-Q for the quarter ended May 5, 2024: https://www.sec.gov/Archives/edgar/data/1474432/000162828024027955/pstg-20240505.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/pure-storage-snowflake-workspace-2024/ --- # Live Nation Entertainment (2024): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Live Nation Entertainment disclosed the incident on 31 May 2024. What was exposed, as the sources state it: The filing says the affected third-party cloud database environment contained Company data, primarily from its Ticketmaster L.L.C. subsidiary, and that a criminal threat actor offered what it alleged to be Company user data for sale on May 27, 2024. - **Disclosed:** 31 May 2024 - **Incident window:** Identified May 20, 2024, as the filing says; start not public - **Cloud or provider:** not public - **What was exposed:** The filing says the affected third-party cloud database environment contained Company data, primarily from its Ticketmaster L.L.C. subsidiary, and that a criminal threat actor offered what it alleged to be Company user data for sale on May 27, 2024. - **How access was gained:** The filing says the Company identified unauthorized activity within a third-party cloud database environment containing Company data; the filing does not say how access was gained. - **Access path:** Not public: the sources do not state how access was gained - **Kind of control that addresses it:** not public (the sources do not state how access was gained) ## In the words of the source “On May 20, 2024, Live Nation Entertainment, Inc. (the Company or we) identified unauthorized activity within a third-party cloud database environment containing Company data” (regulator filing, [Live Nation Form 8-K, Item 8.01, filed 2024-05-31](https://www.sec.gov/Archives/edgar/data/1335258/000133525824000081/lyv-20240520.htm)) ## Primary sources - regulator filing: [Live Nation Form 8-K, Item 8.01, filed 2024-05-31](https://www.sec.gov/Archives/edgar/data/1335258/000133525824000081/lyv-20240520.htm), dated 31 May 2024 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Live%20Nation%20Entertainment%20(live-nation-ticketmaster-2024)) with the entry name in the subject. ## Related pages - [Hugging Face 2024: huggingface spaces secrets](/incidents/huggingface-spaces-secrets-2024/) - [Snowflake customer accounts (multiple organizations) 2024:](/incidents/snowflake-customer-accounts-2024/) - [Dropbox 2024: sign](/incidents/dropbox-sign-2024/) ## Sources - Live Nation Form 8-K, Item 8.01, filed 2024-05-31: https://www.sec.gov/Archives/edgar/data/1335258/000133525824000081/lyv-20240520.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/live-nation-ticketmaster-2024/ --- # Hugging Face (2024): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Hugging Face disclosed the incident on 31 May 2024. What was exposed, as the sources state it: The company stated it suspects a subset of Spaces secrets could have been accessed without authorization, and that it revoked a number of HF tokens present in those secrets. The number of secrets or users is not public. - **Disclosed:** 31 May 2024 - **Incident window:** Detected in the week of 2024-05-27 (the post: "earlier this week") - **Cloud or provider:** Hugging Face - **What was exposed:** The company stated it suspects a subset of Spaces secrets could have been accessed without authorization, and that it revoked a number of HF tokens present in those secrets. The number of secrets or users is not public. - **How access was gained:** The post says the company detected unauthorized access to its Spaces platform, specifically related to Spaces secrets. How access was gained is not public. The company stated it removed organization tokens, added key management for Spaces secrets and expanded detection of leaked tokens. - **Access path:** Not public: the sources do not state how access was gained - **Kind of control that addresses it:** not public (the sources do not state how access was gained) ## In the words of the source “Earlier this week our team detected unauthorized access to our Spaces platform, specifically related to Spaces secrets.” (company statement, [Space secrets leak disclosure (Hugging Face blog)](https://huggingface.co/blog/space-secrets-disclosure)) ## Primary sources - company statement: [Space secrets leak disclosure (Hugging Face blog)](https://huggingface.co/blog/space-secrets-disclosure), dated 31 May 2024 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Hugging%20Face%20(huggingface-spaces-secrets-2024)) with the entry name in the subject. ## Related pages - [Snowflake customer accounts (multiple organizations) 2024:](/incidents/snowflake-customer-accounts-2024/) - [Dropbox 2024: sign](/incidents/dropbox-sign-2024/) - [Payroll2U Pte. Ltd. 2024: pdpc](/incidents/payroll2u-pdpc-2022/) ## Sources - Space secrets leak disclosure (Hugging Face blog): https://huggingface.co/blog/space-secrets-disclosure (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/huggingface-spaces-secrets-2024/ --- # Snowflake customer accounts (multiple organizations) (2024): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Snowflake customer accounts (multiple organizations) disclosed the incident on 30 May 2024. What was exposed, as the sources state it: not public. The ledger reads the access path as stolen user credentials and session tokens. The cloud or provider involved: Snowflake. - **Disclosed:** 30 May 2024 - **Incident window:** from May 2024; the 10-K says further attacks followed - **Cloud or provider:** Snowflake - **What was exposed:** not public - **How access was gained:** Snowflake stated in its annual report that a threat actor accessed a number of customers’ Snowflake accounts because those customers had not met their obligations to implement multi-factor authentication and network access policies, and that it found no breach of its own platform or environment. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** require phishing-resistant multi-factor authentication and network allow-lists for every identity with access to the data platform, and rotate static credentials ## In the words of the source “we became aware that a cybersecurity threat actor had accessed a number of our customers’ Snowflake accounts as a result of such customers’ failure to fulfill certain of their obligations under our shared responsibility cybersecurity model” (regulator filing, [Snowflake Form 10-K for fiscal year ended January 31, 2025](https://www.sec.gov/Archives/edgar/data/1640147/000164014725000052/snow-20250131.htm)) ## Primary sources - regulator filing: [Snowflake Form 10-K for fiscal year ended January 31, 2025](https://www.sec.gov/Archives/edgar/data/1640147/000164014725000052/snow-20250131.htm), dated not stated If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Snowflake%20customer%20accounts%20(multiple%20organizations)%20(snowflake-customer-accounts-2024)) with the entry name in the subject. ## Related pages - [Dropbox 2024: sign](/incidents/dropbox-sign-2024/) - [Payroll2U Pte. Ltd. 2024: pdpc](/incidents/payroll2u-pdpc-2022/) - [Cloudflare 2024: thanksgiving](/incidents/cloudflare-thanksgiving-2023/) ## Sources - Snowflake Form 10-K for fiscal year ended January 31, 2025: https://www.sec.gov/Archives/edgar/data/1640147/000164014725000052/snow-20250131.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/snowflake-customer-accounts-2024/ --- # Dropbox (2024): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Dropbox disclosed the incident on 1 May 2024. The ledger reads the access path as over-permissioned identities with leaked access keys and secrets. The cloud or provider involved: not public. Every field rests on the primary sources listed below. Fields no primary source states read not public. - **Disclosed:** 1 May 2024 - **Incident window:** First access April 19, 2024; last observed activity April 20, 2024, as the company states - **Cloud or provider:** not public - **What was exposed:** The company stated the threat actor accessed Dropbox Sign customer information such as email addresses, user names, phone numbers and hashed passwords, general account settings, and authentication information such as API keys, OAuth tokens and multi-factor authentication; it stated no unauthorized access to document contents or payment information was found. - **How access was gained:** The company stated a third party gained access to an automated system configuration tool, compromised a back-end service account with privileges across the production environment, and reached the customer database; it answered how access was gained with a compromised access token. - **Access path:** [Over-permissioned identities](/access-paths/over-permissioned-identity/); also [Leaked access keys and secrets](/access-paths/leaked-access-key/) - **Kind of control that addresses it:** Scope service accounts to the minimum production actions their automation needs, and protect and rotate the tokens that authenticate them ## In the words of the source “The actor compromised a service account that was part of Sign’s back-end, which is a type of non-human account used to execute applications and run automated services.” (company statement, [A recent security incident involving Dropbox Sign](https://sign.dropbox.com/blog/a-recent-security-incident-involving-dropbox-sign)) ## Primary sources - company statement: [A recent security incident involving Dropbox Sign](https://sign.dropbox.com/blog/a-recent-security-incident-involving-dropbox-sign), dated 21 Jun 2024 - regulator filing: [Dropbox Form 8-K, Item 1.05, filed 2024-05-01](https://www.sec.gov/Archives/edgar/data/1467623/000146762324000024/dbx-20240429.htm), dated 1 May 2024 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Dropbox%20(dropbox-sign-2024)) with the entry name in the subject. ## Related pages - [Payroll2U Pte. Ltd. 2024: pdpc](/incidents/payroll2u-pdpc-2022/) - [Cloudflare 2024: thanksgiving](/incidents/cloudflare-thanksgiving-2023/) - [Hewlett Packard Enterprise 2024: hpe midnight blizzard ema](/incidents/hpe-midnight-blizzard-email-2024/) ## Sources - A recent security incident involving Dropbox Sign: https://sign.dropbox.com/blog/a-recent-security-incident-involving-dropbox-sign (read 16 Nov 2026) - Dropbox Form 8-K, Item 1.05, filed 2024-05-01: https://www.sec.gov/Archives/edgar/data/1467623/000146762324000024/dbx-20240429.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/dropbox-sign-2024/ --- # Payroll2U Pte. Ltd. (2024): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Payroll2U Pte. Ltd. disclosed the incident on 22 Apr 2024. What was exposed, as the sources state it: The PDPC decision states 81.95 GB of data was exfiltrated and posted on the dark web, including the personal data of 5,640 employees of a client: name, bank account number, salary, identity card number, address, date of birth and email. - **Disclosed:** 22 Apr 2024 - **Incident window:** 29 December 2022 to 16 January 2023 - **Cloud or provider:** AWS - **What was exposed:** The PDPC decision states 81.95 GB of data was exfiltrated and posted on the dark web, including the personal data of 5,640 employees of a client: name, bank account number, salary, identity card number, address, date of birth and email. - **How access was gained:** The decision says a single compromised user account was used for Remote Desktop Protocol access to five servers in the company's AWS environment, and that investigators could not determine how the credentials were obtained. It lists no MFA for administrator accounts or accounts with access to personal data. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/); also [Exposed network services and request paths](/access-paths/exposed-network-service/) - **Kind of control that addresses it:** require MFA for every account with access to cloud servers or personal data, limit remote desktop exposure, and detect unusual sign-ins ## In the words of the source “Investigations revealed that unauthorised activity had occurred from 29 December 2022 to 16 January 2023, with a single compromised account used for Remote Desktop Protocol (RDP) access to five servers on the Organisation’s AWS environment.” (regulator filing, [PDPC decision: Payroll2U Pte. Ltd.](https://www.pdpc.gov.sg/assets/f3d9bace-cb3d-4e01-ab1c-5099c882744e)) ## Primary sources - regulator filing: [PDPC decision: Payroll2U Pte. Ltd.](https://www.pdpc.gov.sg/assets/f3d9bace-cb3d-4e01-ab1c-5099c882744e), dated 22 Apr 2024 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Payroll2U%20Pte.%20Ltd.%20(payroll2u-pdpc-2022)) with the entry name in the subject. ## Related pages - [Cloudflare 2024: thanksgiving](/incidents/cloudflare-thanksgiving-2023/) - [Hewlett Packard Enterprise 2024: hpe midnight blizzard ema](/incidents/hpe-midnight-blizzard-email-2024/) - [Microsoft 2024: midnight blizzard](/incidents/microsoft-midnight-blizzard-2024/) ## Sources - PDPC decision: Payroll2U Pte. Ltd.: https://www.pdpc.gov.sg/assets/f3d9bace-cb3d-4e01-ab1c-5099c882744e (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/payroll2u-pdpc-2022/ --- # Cloudflare (2024): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Cloudflare disclosed the incident on 1 Feb 2024. What was exposed, as the sources state it: The company stated the threat actor accessed some documentation and a limited amount of source code on its self-hosted Atlassian server, and that the only production systems reachable with the stolen credentials were its Atlassian environment. - **Disclosed:** 1 Feb 2024 - **Incident window:** Reconnaissance from November 14, 2023; detected November 23, 2023; access terminated November 24, 2023, as the company states - **Cloud or provider:** Okta - **What was exposed:** The company stated the threat actor accessed some documentation and a limited amount of source code on its self-hosted Atlassian server, and that the only production systems reachable with the stolen credentials were its Atlassian environment. - **How access was gained:** The company stated the actor used one access token and three service account credentials taken in the October 2023 Okta compromise that the company had failed to rotate. - **Access path:** [Leaked access keys and secrets](/access-paths/leaked-access-key/); also [Third-party access](/access-paths/third-party-access/) - **Kind of control that addresses it:** Rotate every credential exposed by an upstream provider incident, and inventory service tokens and service accounts so none are assumed unused ## In the words of the source “They did this by using one access token and three service account credentials that had been taken, and that we failed to rotate, after the Okta compromise of October 2023 .” (company statement, [Thanksgiving 2023 security incident](https://blog.cloudflare.com/thanksgiving-2023-security-incident/)) ## Primary sources - company statement: [Thanksgiving 2023 security incident](https://blog.cloudflare.com/thanksgiving-2023-security-incident/), dated 1 Feb 2024 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Cloudflare%20(cloudflare-thanksgiving-2023)) with the entry name in the subject. ## Related pages - [Hewlett Packard Enterprise 2024: hpe midnight blizzard ema](/incidents/hpe-midnight-blizzard-email-2024/) - [Microsoft 2024: midnight blizzard](/incidents/microsoft-midnight-blizzard-2024/) - [A bank headquartered in San Francisco (not named in the U.](/incidents/sf-bank-cloud-engineer-retained-laptop-2020/) ## Sources - Thanksgiving 2023 security incident: https://blog.cloudflare.com/thanksgiving-2023-security-incident/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/cloudflare-thanksgiving-2023/ --- # Hewlett Packard Enterprise (2024): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Hewlett Packard Enterprise disclosed the incident on 24 Jan 2024. The sources do not state how access was gained, so the access path reads not public. The cloud or provider involved: not public. Every field rests on the primary sources listed below. - **Disclosed:** 24 Jan 2024 - **Incident window:** Data accessed and exfiltrated from May 2023; the company was notified on December 12, 2023 (per the filing) - **Cloud or provider:** not public - **What was exposed:** The filing says the threat actor accessed and exfiltrated data from a small percentage of HPE mailboxes belonging to individuals in cybersecurity, go-to-market, business segments and other functions, and earlier a limited number of SharePoint files. The provider of the cloud-based email environment is not named in the filing. - **How access was gained:** The filing says a suspected nation-state actor, which the company believes to be the threat actor Midnight Blizzard, gained unauthorized access to HPE's cloud-based email environment. How access was gained is not public. - **Access path:** Not public: the sources do not state how access was gained - **Kind of control that addresses it:** not public (the sources do not state how access was gained) ## In the words of the source “a suspected nation-state actor, believed to be the threat actor Midnight Blizzard, the state-sponsored actor also known as Cozy Bear, had gained unauthorized access to HPE’s cloud-based email environment” (regulator filing, [Hewlett Packard Enterprise Form 8-K, Item 1.05, filed 2024-01-24](https://www.sec.gov/Archives/edgar/data/1645590/000164559024000009/hpe-20240119.htm)) ## Primary sources - regulator filing: [Hewlett Packard Enterprise Form 8-K, Item 1.05, filed 2024-01-24](https://www.sec.gov/Archives/edgar/data/1645590/000164559024000009/hpe-20240119.htm), dated 24 Jan 2024 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Hewlett%20Packard%20Enterprise%20(hpe-midnight-blizzard-email-2024)) with the entry name in the subject. ## Related pages - [Microsoft 2024: midnight blizzard](/incidents/microsoft-midnight-blizzard-2024/) - [A bank headquartered in San Francisco (not named in the U.](/incidents/sf-bank-cloud-engineer-retained-laptop-2020/) - [Global Tel*Link Corp. 2023: ftc](/incidents/global-tel-link-ftc-2020/) ## Sources - Hewlett Packard Enterprise Form 8-K, Item 1.05, filed 2024-01-24: https://www.sec.gov/Archives/edgar/data/1645590/000164559024000009/hpe-20240119.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/hpe-midnight-blizzard-email-2024/ --- # Microsoft (2024): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Microsoft disclosed the incident on 19 Jan 2024. What was exposed, as the sources state it: Microsoft stated that some emails and attached documents were exfiltrated from a very small percentage of corporate email accounts, including senior leadership and cybersecurity and legal staff; a March 2024 update says some source code repositories and internal systems were also accessed. - **Disclosed:** 19 Jan 2024 - **Incident window:** Late November 2023 to January 12, 2024 (detection), as the company states - **Cloud or provider:** Microsoft 365 - **What was exposed:** Microsoft stated that some emails and attached documents were exfiltrated from a very small percentage of corporate email accounts, including senior leadership and cybersecurity and legal staff; a March 2024 update says some source code repositories and internal systems were also accessed. - **How access was gained:** Microsoft stated the threat actor used a password spray attack to compromise a legacy non-production test tenant account, then used that account's permissions to access corporate email accounts. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/); also [Over-permissioned identities](/access-paths/over-permissioned-identity/) - **Kind of control that addresses it:** Apply phishing-resistant MFA and lockout to every tenant account including legacy and test tenants, and scope application permissions so a test-tenant identity cannot reach production mailboxes ## In the words of the source “Beginning in late November 2023, the threat actor used a password spray attack to compromise a legacy non-production test tenant account and gain a foothold” (provider post mortem, [Microsoft Actions Following Attack by Nation State Actor Midnight Blizzard](https://www.microsoft.com/en-us/msrc/blog/2024/01/microsoft-actions-following-attack-by-nation-state-actor-midnight-blizzard)) ## Primary sources - provider post mortem: [Microsoft Actions Following Attack by Nation State Actor Midnight Blizzard](https://www.microsoft.com/en-us/msrc/blog/2024/01/microsoft-actions-following-attack-by-nation-state-actor-midnight-blizzard), dated 19 Jan 2024 - provider post mortem: [Update on Microsoft Actions Following Attack by Nation State Actor Midnight Blizzard](https://www.microsoft.com/en-us/msrc/blog/2024/03/update-on-microsoft-actions-following-attack-by-nation-state-actor-midnight-blizzard), dated 8 Mar 2024 - regulator filing: [Microsoft Form 8-K, Item 1.05, filed 2024-01-19](https://www.sec.gov/Archives/edgar/data/789019/000119312524011295/d708866d8k.htm), dated 19 Jan 2024 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Microsoft%20(microsoft-midnight-blizzard-2024)) with the entry name in the subject. ## Related pages - [A bank headquartered in San Francisco (not named in the U.](/incidents/sf-bank-cloud-engineer-retained-laptop-2020/) - [Global Tel*Link Corp. 2023: ftc](/incidents/global-tel-link-ftc-2020/) - [Sumo Logic 2023: aws credential](/incidents/sumo-logic-aws-credential-2023/) ## Sources - Microsoft Actions Following Attack by Nation State Actor Midnight Blizzard: https://www.microsoft.com/en-us/msrc/blog/2024/01/microsoft-actions-following-attack-by-nation-state-actor-midnight-blizzard (read 16 Nov 2026) - Update on Microsoft Actions Following Attack by Nation State Actor Midnight Blizzard: https://www.microsoft.com/en-us/msrc/blog/2024/03/update-on-microsoft-actions-following-attack-by-nation-state-actor-midnight-blizzard (read 16 Nov 2026) - Microsoft Form 8-K, Item 1.05, filed 2024-01-19: https://www.sec.gov/Archives/edgar/data/789019/000119312524011295/d708866d8k.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/microsoft-midnight-blizzard-2024/ --- # A bank headquartered in San Francisco (not named in the U.S. Attorney release) (2023): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. A bank headquartered in San Francisco (not named in the U.S. Attorney release) disclosed the incident on 11 Dec 2023. The ledger reads the access path as retained access after departure or change. The cloud or provider involved: not public. - **Disclosed:** 11 Dec 2023 - **Incident window:** Night of March 11, 2020 into the following morning (per the indictment as summarised in the release) - **Cloud or provider:** not public - **What was exposed:** The release says the bank's code repositories were deleted, a malicious script deleted logs, taunts were left in the code, sessions were opened in other employees' names, and proprietary bank code was emailed out. The release's subheading describes damage to the bank's cloud system. The cloud provider is not named. - **How access was gained:** The release says a cloud engineer fired on March 11, 2020 used a company-issued laptop he had not returned to access the bank's computer network without authorization. - **Access path:** [Retained access after departure or change](/access-paths/retained-access/) - **Kind of control that addresses it:** revoke all cloud console, repository and VPN access and recover or remotely wipe company devices at the moment of termination, and alert on privileged activity by deprovisioned identities ## In the words of the source “used his company-issued laptop—which he failed to return upon being fired—to access the bank’s computer network without authorization and to cause substantial damage.” (court document, [Disgruntled Cloud Engineer Sentenced To Two Years In Prison (U.S. Attorney, Northern District of California), archived copy](https://web.archive.org/web/2026/https://www.justice.gov/usao-ndca/pr/disgruntled-cloud-engineer-sentenced-two-years-prison-intentionally-damaging-his)) ## Primary sources - court document: [Disgruntled Cloud Engineer Sentenced To Two Years In Prison (U.S. Attorney, Northern District of California), archived copy](https://web.archive.org/web/2026/https://www.justice.gov/usao-ndca/pr/disgruntled-cloud-engineer-sentenced-two-years-prison-intentionally-damaging-his), dated 11 Dec 2023 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20A%20bank%20headquartered%20in%20San%20Francisco%20(not%20named%20in%20the%20U.S.%20Attorney%20release)%20(sf-bank-cloud-engineer-retained-laptop-2020)) with the entry name in the subject. ## Related pages - [Global Tel*Link Corp. 2023: ftc](/incidents/global-tel-link-ftc-2020/) - [Sumo Logic 2023: aws credential](/incidents/sumo-logic-aws-credential-2023/) - [1Password 2023: okta](/incidents/1password-okta-2023/) ## Sources - Disgruntled Cloud Engineer Sentenced To Two Years In Prison (U.S. Attorney, Northern District of California), archived copy: https://web.archive.org/web/2026/https://www.justice.gov/usao-ndca/pr/disgruntled-cloud-engineer-sentenced-two-years-prison-intentionally-damaging-his (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/sf-bank-cloud-engineer-retained-laptop-2020/ --- # Global Tel*Link Corp. (2023): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Global Tel*Link Corp. disclosed the incident on 16 Nov 2023. The ledger reads the access path as public storage, databases and services with third-party access. The cloud or provider involved: not public. Every field rests on the primary sources listed below. - **Disclosed:** 16 Nov 2023 - **Incident window:** August 2020 to early September 2020 (the FTC press release says data was copied to the cloud in August 2020 and that the company was notified of data on the dark web in early September) - **Cloud or provider:** not public - **What was exposed:** The FTC states the personal information of nearly 650,000 real users, including full names, dates of birth, phone numbers and usernames, was copied unencrypted into a cloud test environment. The FTC says a forensic analysis showed that a handful of hackers accessed billions of bytes of the exposed data. - **How access was gained:** The FTC press release says that, after the company and a third-party vendor copied the data into the cloud for testing, changes made by the vendor to the security settings left the data accessible via the internet without safeguards, until a security researcher alerted the company. - **Access path:** [Public storage, databases and services](/access-paths/public-resource/); also [Third-party access](/access-paths/third-party-access/) - **Kind of control that addresses it:** require review and approval of any security-setting change on cloud environments that hold personal data, keep real personal data out of test environments, and alert on settings that expose a data store to the internet ## In the words of the source “As a result of changes made by the company’s third-party vendor to the security settings for the data stored in the cloud, the personal data of many Global Tel*Link customers was left accessible via the internet” (regulator filing, [FTC press release: FTC Takes Action Against Global Tel*Link Corp. for Failing to Adequately Secure Data, Notify Consumers After Their Personal Data Was Breached (web.archive.org copy)](https://web.archive.org/web/2024/https://www.ftc.gov/news-events/news/press-releases/2023/11/ftc-takes-action-against-global-tellink-corp-failing-adequately-secure-data-notify-consumers-after)) ## Primary sources - regulator filing: [FTC press release: FTC Takes Action Against Global Tel*Link Corp. for Failing to Adequately Secure Data, Notify Consumers After Their Personal Data Was Breached (web.archive.org copy)](https://web.archive.org/web/2024/https://www.ftc.gov/news-events/news/press-releases/2023/11/ftc-takes-action-against-global-tellink-corp-failing-adequately-secure-data-notify-consumers-after), dated 16 Nov 2023 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Global%20Tel*Link%20Corp.%20(global-tel-link-ftc-2020)) with the entry name in the subject. ## Related pages - [Sumo Logic 2023: aws credential](/incidents/sumo-logic-aws-credential-2023/) - [1Password 2023: okta](/incidents/1password-okta-2023/) - [Okta 2023: support system](/incidents/okta-support-system-2023/) ## Sources - FTC press release: FTC Takes Action Against Global Tel*Link Corp. for Failing to Adequately Secure Data, Notify Consumers After Their Personal Data Was Breached (web.archive.org copy): https://web.archive.org/web/2024/https://www.ftc.gov/news-events/news/press-releases/2023/11/ftc-takes-action-against-global-tellink-corp-failing-adequately-secure-data-notify-consumers-after (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/global-tel-link-ftc-2020/ --- # Sumo Logic (2023): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Sumo Logic disclosed the incident on 7 Nov 2023. What was exposed, as the sources state it: The company stated that, after investigation by its teams and third-party forensic experts, it found no proof of customer data impact. It advised customers to rotate API access keys and stored third-party credentials. - **Disclosed:** 7 Nov 2023 - **Incident window:** Discovered November 3, 2023 - **Cloud or provider:** AWS - **What was exposed:** The company stated that, after investigation by its teams and third-party forensic experts, it found no proof of customer data impact. It advised customers to rotate API access keys and stored third-party credentials. - **How access was gained:** The customer notice says the activity used a compromised credential to access a Sumo Logic AWS account. How the credential was compromised is not public. - **Access path:** [Leaked access keys and secrets](/access-paths/leaked-access-key/) - **Kind of control that addresses it:** replace long-lived credentials for cloud accounts with short-lived role credentials and alert on their use from new locations ## In the words of the source “The activity identified used a compromised credential to access a Sumo Logic AWS account.” (company statement, [Security response center (Sumo Logic), archived copy with the November 2023 incident notices](https://web.archive.org/web/20240101000000/https://www.sumologic.com/security-response-center/)) ## Primary sources - company statement: [Security response center (Sumo Logic), archived copy with the November 2023 incident notices](https://web.archive.org/web/20240101000000/https://www.sumologic.com/security-response-center/), dated 7 Nov 2023 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Sumo%20Logic%20(sumo-logic-aws-credential-2023)) with the entry name in the subject. ## Related pages - [1Password 2023: okta](/incidents/1password-okta-2023/) - [Okta 2023: support system](/incidents/okta-support-system-2023/) - [BeyondTrust 2023: okta](/incidents/beyondtrust-okta-2023/) ## Sources - Security response center (Sumo Logic), archived copy with the November 2023 incident notices: https://web.archive.org/web/20240101000000/https://www.sumologic.com/security-response-center/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/sumo-logic-aws-credential-2023/ --- # 1Password (2023): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. 1Password disclosed the incident on 23 Oct 2023. What was exposed, as the sources state it: The post says no 1Password user data was accessed and no compromise of user data or other sensitive systems was found. The ledger reads the access path as third-party access. - **Disclosed:** 23 Oct 2023 - **Incident window:** September 29 to October 2023, as the post says - **Cloud or provider:** Okta - **What was exposed:** The post says no 1Password user data was accessed and no compromise of user data or other sensitive systems was found. - **How access was gained:** The company stated it detected suspicious activity on its Okta instance used to manage employee-facing apps on September 29, and later confirmed it was a result of the provider’s Support System breach. - **Access path:** [Third-party access](/access-paths/third-party-access/) - **Kind of control that addresses it:** treat identity-provider support exchanges as sensitive, strip session material from files sent to a provider, and alert on administrator sessions that originate outside known networks ## In the words of the source “As of late Friday, October 20, we’ve confirmed that this was a result of Okta’s Support System breach” (company statement, [Okta Support System incident and 1Password](https://blog.1password.com/okta-incident/)) ## Primary sources - company statement: [Okta Support System incident and 1Password](https://blog.1password.com/okta-incident/), dated 23 Oct 2023 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%201Password%20(1password-okta-2023)) with the entry name in the subject. ## Related pages - [Okta 2023: support system](/incidents/okta-support-system-2023/) - [BeyondTrust 2023: okta](/incidents/beyondtrust-okta-2023/) - [23andMe 2023: stolen user credentials and session tokens](/incidents/23andme-2023/) ## Sources - Okta Support System incident and 1Password: https://blog.1password.com/okta-incident/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/1password-okta-2023/ --- # Okta (2023): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Okta disclosed the incident on 20 Oct 2023. What was exposed, as the sources state it: Okta stated that files associated with 134 customers were accessed in its customer support system; some were HAR files containing session tokens, and the threat actor used them to hijack the sessions of 5 customers. - **Disclosed:** 20 Oct 2023 - **Incident window:** September 28, 2023 to October 17, 2023, as the company states - **Cloud or provider:** Okta - **What was exposed:** Okta stated that files associated with 134 customers were accessed in its customer support system; some were HAR files containing session tokens, and the threat actor used them to hijack the sessions of 5 customers. - **How access was gained:** Okta stated the access used a service account stored in the support system itself, whose username and password had been saved in an employee's personal Google account on an Okta-managed laptop. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** Block sign-in to personal profiles on managed browsers, bind admin session tokens to network location, and require customers' support uploads to be sanitised of session tokens ## In the words of the source “The username and password of the service account had been saved into the employee’s personal Google account.” (provider post mortem, [Unauthorized Access to Okta's Support Case Management System: Root Cause and Remediation](https://sec.okta.com/articles/2023/11/unauthorized-access-oktas-support-case-management-system-root-cause/)) ## Primary sources - provider post mortem: [Unauthorized Access to Okta's Support Case Management System: Root Cause and Remediation](https://sec.okta.com/articles/2023/11/unauthorized-access-oktas-support-case-management-system-root-cause/), dated 3 Nov 2023 - company statement: [Tracking Unauthorized Access to Okta's Support System](https://sec.okta.com/articles/2023/10/tracking-unauthorized-access-oktas-support-system/), dated 20 Oct 2023 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Okta%20(okta-support-system-2023)) with the entry name in the subject. ## Related pages - [BeyondTrust 2023: okta](/incidents/beyondtrust-okta-2023/) - [23andMe 2023: stolen user credentials and session tokens](/incidents/23andme-2023/) - [Microsoft 2023: ai sas token](/incidents/microsoft-ai-sas-token-2023/) ## Sources - Unauthorized Access to Okta's Support Case Management System: Root Cause and Remediation: https://sec.okta.com/articles/2023/11/unauthorized-access-oktas-support-case-management-system-root-cause/ (read 16 Nov 2026) - Tracking Unauthorized Access to Okta's Support System: https://sec.okta.com/articles/2023/10/tracking-unauthorized-access-oktas-support-system/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/okta-support-system-2023/ --- # BeyondTrust (2023): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. BeyondTrust disclosed the incident on 20 Oct 2023. What was exposed, as the sources state it: The post says there was no impact or exposure to BeyondTrust’s infrastructure or customers. The ledger reads the access path as third-party access with stolen user credentials and session tokens. - **Disclosed:** 20 Oct 2023 - **Incident window:** October 2, 2023, as the post says - **Cloud or provider:** Okta - **What was exposed:** The post says there was no impact or exposure to BeyondTrust’s infrastructure or customers. - **How access was gained:** The company stated it detected an attacker trying to access an in-house Okta administrator account using a valid session cookie stolen from the provider’s support system. - **Access path:** [Third-party access](/access-paths/third-party-access/); also [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** bind administrator sessions to the device and network they started on, restrict identity-provider admin access to managed networks, and alert on session reuse from new locations ## In the words of the source “The incident began when BeyondTrust security teams detected an attacker trying to access an in-house Okta administrator account using a valid session cookie stolen from Okta’s support system.” (company statement, [BeyondTrust Discovers Breach of Okta Support Unit](https://www.beyondtrust.com/blog/entry/okta-support-unit-breach)) ## Primary sources - company statement: [BeyondTrust Discovers Breach of Okta Support Unit](https://www.beyondtrust.com/blog/entry/okta-support-unit-breach), dated 20 Oct 2023 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20BeyondTrust%20(beyondtrust-okta-2023)) with the entry name in the subject. ## Related pages - [23andMe 2023: stolen user credentials and session tokens](/incidents/23andme-2023/) - [Microsoft 2023: ai sas token](/incidents/microsoft-ai-sas-token-2023/) - [Sourcegraph 2023: leaked access keys and secrets](/incidents/sourcegraph-2023/) ## Sources - BeyondTrust Discovers Breach of Okta Support Unit: https://www.beyondtrust.com/blog/entry/okta-support-unit-breach (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/beyondtrust-okta-2023/ --- # 23andMe (2023): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. 23andMe disclosed the incident on 6 Oct 2023. What was exposed, as the sources state it: The company stated the threat actor accessed roughly 14,000 user accounts (less than 0.1% of 14 million customers) and, through them, information in approximately 5.5 million DNA Relatives profiles and approximately 1.4 million Family Tree profiles. - **Disclosed:** 6 Oct 2023 - **Incident window:** not public - **Cloud or provider:** not public - **What was exposed:** The company stated the threat actor accessed roughly 14,000 user accounts (less than 0.1% of 14 million customers) and, through them, information in approximately 5.5 million DNA Relatives profiles and approximately 1.4 million Family Tree profiles. - **How access was gained:** The company stated the threat actor accessed accounts through credential stuffing, using usernames and passwords that were also used on other websites previously compromised. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** Require a second authentication factor for every consumer account and screen new passwords against known-compromised credential lists ## In the words of the source “That is, usernames and passwords that were used on 23andMe.com were the same as those used on other websites that have been previously compromised or otherwise available.” (company statement, [Addressing Data Security Concerns - Action Plan](https://blog.23andme.com/articles/addressing-data-security-concerns)) ## Primary sources - company statement: [Addressing Data Security Concerns - Action Plan](https://blog.23andme.com/articles/addressing-data-security-concerns), dated 6 Oct 2023 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%2023andMe%20(23andme-2023)) with the entry name in the subject. ## Related pages - [Microsoft 2023: ai sas token](/incidents/microsoft-ai-sas-token-2023/) - [Sourcegraph 2023: leaked access keys and secrets](/incidents/sourcegraph-2023/) - [Retool 2023: stolen user credentials and session tokens](/incidents/retool-2023/) ## Sources - Addressing Data Security Concerns - Action Plan: https://blog.23andme.com/articles/addressing-data-security-concerns (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/23andme-2023/ --- # Microsoft (2023): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Microsoft disclosed the incident on 18 Sep 2023. What was exposed, as the sources state it: Microsoft stated that data in an internal storage account included backups of two former employees' workstation profiles and internal Microsoft Teams messages of those employees; it stated no customer data was exposed. - **Disclosed:** 18 Sep 2023 - **Incident window:** not public - **Cloud or provider:** Microsoft Azure - **What was exposed:** Microsoft stated that data in an internal storage account included backups of two former employees' workstation profiles and internal Microsoft Teams messages of those employees; it stated no customer data was exposed. - **How access was gained:** Microsoft stated an employee shared a URL for a blob store in a public GitHub repository, and the URL included an overly-permissive Shared Access Signature token for an internal storage account. - **Access path:** [Leaked access keys and secrets](/access-paths/leaked-access-key/); also [Over-permissioned identities](/access-paths/over-permissioned-identity/) - **Kind of control that addresses it:** Issue storage access tokens with narrow scope and short expiry, and scan public code repositories for them before and after publication ## In the words of the source “This URL included an overly-permissive Shared Access Signature (SAS) token for an internal storage account.” (provider post mortem, [Microsoft mitigated exposure of internal information in a storage account due to overly-permissive SAS token](https://www.microsoft.com/en-us/msrc/blog/2023/09/microsoft-mitigated-exposure-of-internal-information-in-a-storage-account-due-to-overly-permissive-sas-token)) ## Primary sources - provider post mortem: [Microsoft mitigated exposure of internal information in a storage account due to overly-permissive SAS token](https://www.microsoft.com/en-us/msrc/blog/2023/09/microsoft-mitigated-exposure-of-internal-information-in-a-storage-account-due-to-overly-permissive-sas-token), dated 18 Sep 2023 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Microsoft%20(microsoft-ai-sas-token-2023)) with the entry name in the subject. ## Related pages - [Sourcegraph 2023: leaked access keys and secrets](/incidents/sourcegraph-2023/) - [Retool 2023: stolen user credentials and session tokens](/incidents/retool-2023/) - [E-Commerce Enablers Pte. Ltd. 2023: ecommerce pdpc](/incidents/ecommerce-enablers-pdpc-2020/) ## Sources - Microsoft mitigated exposure of internal information in a storage account due to overly-permissive SAS token: https://www.microsoft.com/en-us/msrc/blog/2023/09/microsoft-mitigated-exposure-of-internal-information-in-a-storage-account-due-to-overly-permissive-sas-token (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/microsoft-ai-sas-token-2023/ --- # Sourcegraph (2023): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Sourcegraph disclosed the incident on 30 Aug 2023. What was exposed, as the sources state it: The company stated that paid customers' license key recipient names and email addresses, a small subset of license keys, and community users' account email addresses may have been accessed; it stated no private code or other customer data was accessible. - **Disclosed:** 30 Aug 2023 - **Incident window:** Token committed July 14, 2023; unauthorized admin access August 30, 2023, as the company states - **Cloud or provider:** not public - **What was exposed:** The company stated that paid customers' license key recipient names and email addresses, a small subset of license keys, and community users' account email addresses may have been accessed; it stated no private code or other customer data was accessible. - **How access was gained:** The company stated a malicious actor used a leaked site-admin access token, accidentally committed in a pull request, to raise an account to site-admin on Sourcegraph.com. - **Access path:** [Leaked access keys and secrets](/access-paths/leaked-access-key/) - **Kind of control that addresses it:** Run secret scanning on every commit with push-time blocking, and issue admin tokens with short expiry and narrow scope ## In the words of the source “Sourcegraph experienced a security incident on August 30, 2023 where a malicious actor used a leaked admin access token in our public Sourcegraph instance at Sourcegraph.com.” (company statement, [Security update: Incident involving unauthorized admin access](https://sourcegraph.com/blog/security-update-august-2023)) ## Primary sources - company statement: [Security update: Incident involving unauthorized admin access](https://sourcegraph.com/blog/security-update-august-2023), dated 30 Aug 2023 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Sourcegraph%20(sourcegraph-2023)) with the entry name in the subject. ## Related pages - [Retool 2023: stolen user credentials and session tokens](/incidents/retool-2023/) - [E-Commerce Enablers Pte. Ltd. 2023: ecommerce pdpc](/incidents/ecommerce-enablers-pdpc-2020/) - [JumpCloud 2023: orchestration](/incidents/jumpcloud-orchestration-2023/) ## Sources - Security update: Incident involving unauthorized admin access: https://sourcegraph.com/blog/security-update-august-2023 (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/sourcegraph-2023/ --- # Retool (2023): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Retool disclosed the incident on 29 Aug 2023. What was exposed, as the sources state it: The company stated that 27 cloud customer accounts were accessed without authorization (account takeovers, with emails changed and passwords reset); it stated no on-prem or managed accounts were affected. - **Disclosed:** 29 Aug 2023 - **Incident window:** Attack on August 27, 2023; affected customers notified August 29, 2023, as the company states - **Cloud or provider:** Okta - **What was exposed:** The company stated that 27 cloud customer accounts were accessed without authorization (account takeovers, with emails changed and passwords reset); it stated no on-prem or managed accounts were affected. - **How access was gained:** The company stated an employee was targeted by an SMS-based phishing attack followed by a call, and that the attacker added their own device to the employee's Okta account and used synced MFA codes to reach the VPN and internal admin systems. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** Require phishing-resistant, device-bound authentication factors for all staff and keep one-time codes out of synchronised cloud accounts ## In the words of the source “The attacker was able to navigate through multiple layers of security controls after taking advantage of one of our employees through a SMS-based phishing attack.” (company statement, [MFA isn't MFA](https://retool.com/blog/mfa-isnt-mfa)) ## Primary sources - company statement: [MFA isn't MFA](https://retool.com/blog/mfa-isnt-mfa), dated not stated If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Retool%20(retool-2023)) with the entry name in the subject. ## Related pages - [E-Commerce Enablers Pte. Ltd. 2023: ecommerce pdpc](/incidents/ecommerce-enablers-pdpc-2020/) - [JumpCloud 2023: orchestration](/incidents/jumpcloud-orchestration-2023/) - [Microsoft 2023: storm 0558](/incidents/microsoft-storm-0558-2023/) ## Sources - MFA isn't MFA: https://retool.com/blog/mfa-isnt-mfa (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/retool-2023/ --- # E-Commerce Enablers Pte. Ltd. (2023): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. E-Commerce Enablers Pte. Ltd. disclosed the incident on 16 Aug 2023. What was exposed, as the sources state it: The PDPC decision states customer data on virtual servers in an AWS environment was exfiltrated, with counts including 1,457,637 email addresses, 840,210 names, 447,076 mobile numbers, 299,381 bank account numbers and 378,531 partial credit card records. - **Disclosed:** 16 Aug 2023 - **Incident window:** 4 June 2019 (key committed) to 9 September 2020 (key used); discovered 17 September 2020 - **Cloud or provider:** AWS - **What was exposed:** The PDPC decision states customer data on virtual servers in an AWS environment was exfiltrated, with counts including 1,457,637 email addresses, 840,210 names, 447,076 mobile numbers, 299,381 bank account numbers and 378,531 partial credit card records. - **How access was gained:** The decision says an AWS access key with full administrative privileges was committed to a private GitHub repository on 4 June 2019, removed on 6 June but still viewable in the commit history, and not fully disabled when a replacement key was created. On 9 September 2020 a threat actor used it to reach the AWS environment, changed security settings to allow remote internet access to database instances and exfiltrated data. The company conjectured the key was found in the commit history. - **Access path:** [Leaked access keys and secrets](/access-paths/leaked-access-key/); also [Over-permissioned identities](/access-paths/over-permissioned-identity/) - **Kind of control that addresses it:** never commit cloud keys to code repositories, scan repository history for secrets, fully revoke replaced keys, and avoid full-administrator keys for routine deployment work ## In the words of the source “On 9 September 2020, a malicious threat actor accessed the Organisation’s AWS environment utilizing the AWS Key.” (regulator filing, [PDPC decision: E-Commerce Enablers Pte. Ltd.](https://www.pdpc.gov.sg/assets/777b18ae-71e5-4e5d-9e89-9d1b54aca926)) ## Primary sources - regulator filing: [PDPC decision: E-Commerce Enablers Pte. Ltd.](https://www.pdpc.gov.sg/assets/777b18ae-71e5-4e5d-9e89-9d1b54aca926), dated 16 Aug 2023 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20E-Commerce%20Enablers%20Pte.%20Ltd.%20(ecommerce-enablers-pdpc-2020)) with the entry name in the subject. ## Related pages - [JumpCloud 2023: orchestration](/incidents/jumpcloud-orchestration-2023/) - [Microsoft 2023: storm 0558](/incidents/microsoft-storm-0558-2023/) - [1Health.io Inc. (formerly Vitagene, Inc.) 2023: ftc](/incidents/vitagene-1health-ftc-2019/) ## Sources - PDPC decision: E-Commerce Enablers Pte. Ltd.: https://www.pdpc.gov.sg/assets/777b18ae-71e5-4e5d-9e89-9d1b54aca926 (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/ecommerce-enablers-pdpc-2020/ --- # JumpCloud (2023): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. JumpCloud disclosed the incident on 12 Jul 2023. What was exposed, as the sources state it: The company stated fewer than 5 customers and fewer than 10 devices were impacted, out of more than 200,000 organizations on the platform, and that all impacted customers were notified directly. - **Disclosed:** 12 Jul 2023 - **Incident window:** June 22 to July 2023 (the company: spear-phishing on June 22, anomalous activity found June 27, customer impact found July 5) - **Cloud or provider:** JumpCloud - **What was exposed:** The company stated fewer than 5 customers and fewer than 10 devices were impacted, out of more than 200,000 organizations on the platform, and that all impacted customers were notified directly. - **How access was gained:** The company stated it traced anomalous activity on an internal orchestration system to a spear-phishing campaign on June 22, which gave unauthorized access to a specific area of its infrastructure. It later found unusual activity in the commands framework for a small set of customers, and stated the attack vector was data injection into its commands framework. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/); also [Cross-tenant access](/access-paths/cross-tenant/) - **Kind of control that addresses it:** require phishing-resistant authentication for staff with access to orchestration systems, and isolate customer command execution from internal tooling ## In the words of the source “On June 27 at 15:13 UTC we discovered anomalous activity on an internal orchestration system which we traced back to a sophisticated spear-phishing campaign perpetrated by the threat actor on June 22.” (company statement, [[Security Update] Incident Details (JumpCloud blog)](https://jumpcloud.com/blog/security-update-incident-details)) ## Primary sources - company statement: [[Security Update] Incident Details (JumpCloud blog)](https://jumpcloud.com/blog/security-update-incident-details), dated 20 Jul 2023 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20JumpCloud%20(jumpcloud-orchestration-2023)) with the entry name in the subject. ## Related pages - [Microsoft 2023: storm 0558](/incidents/microsoft-storm-0558-2023/) - [1Health.io Inc. (formerly Vitagene, Inc.) 2023: ftc](/incidents/vitagene-1health-ftc-2019/) - [GitHub 2023: rsa ssh key](/incidents/github-rsa-ssh-key-2023/) ## Sources - [Security Update] Incident Details (JumpCloud blog): https://jumpcloud.com/blog/security-update-incident-details (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/jumpcloud-orchestration-2023/ --- # Microsoft (2023): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Microsoft disclosed the incident on 11 Jul 2023. What was exposed, as the sources state it: Microsoft stated that email accounts affecting approximately 25 organizations in the public cloud, including government agencies, and related consumer accounts were accessed. The ledger reads the access path as cross-tenant access with leaked access keys and secrets. - **Disclosed:** 11 Jul 2023 - **Incident window:** From May 15, 2023, as the company states - **Cloud or provider:** Microsoft 365 - **What was exposed:** Microsoft stated that email accounts affecting approximately 25 organizations in the public cloud, including government agencies, and related consumer accounts were accessed. - **How access was gained:** Microsoft stated the actor used an acquired Microsoft account consumer signing key to forge tokens, and exploited a token validation issue so that enterprise mail accepted them. The post-mortem says the most probable mechanism for the key acquisition was a compromised engineer corporate account with access to a debugging environment holding a crash dump that contained the key. - **Access path:** [Cross-tenant access](/access-paths/cross-tenant/); also [Leaked access keys and secrets](/access-paths/leaked-access-key/) - **Kind of control that addresses it:** Validate token issuer and key scope in every service that accepts signed tokens, and keep signing key material inside isolated key-management systems with credential scanning of debug artefacts ## In the words of the source “The actor exploited a token validation issue to impersonate Azure AD users and gain access to enterprise mail.” (provider post mortem, [Microsoft mitigates China-based threat actor Storm-0558 targeting of customer email](https://www.microsoft.com/en-us/msrc/blog/2023/07/microsoft-mitigates-china-based-threat-actor-storm-0558-targeting-of-customer-email)) ## Primary sources - provider post mortem: [Microsoft mitigates China-based threat actor Storm-0558 targeting of customer email](https://www.microsoft.com/en-us/msrc/blog/2023/07/microsoft-mitigates-china-based-threat-actor-storm-0558-targeting-of-customer-email), dated 11 Jul 2023 - provider post mortem: [Results of Major Technical Investigations for Storm-0558 Key Acquisition](https://www.microsoft.com/en-us/msrc/blog/2023/09/results-of-major-technical-investigations-for-storm-0558-key-acquisition), dated 6 Sep 2023 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Microsoft%20(microsoft-storm-0558-2023)) with the entry name in the subject. ## Related pages - [1Health.io Inc. (formerly Vitagene, Inc.) 2023: ftc](/incidents/vitagene-1health-ftc-2019/) - [GitHub 2023: rsa ssh key](/incidents/github-rsa-ssh-key-2023/) - [Eatigo International Pte. Ltd. 2023: pdpc](/incidents/eatigo-pdpc-2020/) ## Sources - Microsoft mitigates China-based threat actor Storm-0558 targeting of customer email: https://www.microsoft.com/en-us/msrc/blog/2023/07/microsoft-mitigates-china-based-threat-actor-storm-0558-targeting-of-customer-email (read 16 Nov 2026) - Results of Major Technical Investigations for Storm-0558 Key Acquisition: https://www.microsoft.com/en-us/msrc/blog/2023/09/results-of-major-technical-investigations-for-storm-0558-key-acquisition (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/microsoft-storm-0558-2023/ --- # 1Health.io Inc. (formerly Vitagene, Inc.) (2023): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. 1Health.io Inc. (formerly Vitagene, Inc.) disclosed the incident on 16 Jun 2023. What was exposed, as the sources state it: The FTC states nearly 2,400 health reports about consumers and raw genetic data of at least 227 consumers, sometimes accompanied by a first name, sat in publicly accessible buckets, unencrypted, according to the complaint. - **Disclosed:** 16 Jun 2023 - **Incident window:** 2019 (a security researcher contacted the company in June 2019; the FTC says the company was warned at least three times over a two-year period) - **Cloud or provider:** AWS - **What was exposed:** The FTC states nearly 2,400 health reports about consumers and raw genetic data of at least 227 consumers, sometimes accompanied by a first name, sat in publicly accessible buckets, unencrypted, according to the complaint. - **How access was gained:** The FTC press release says the company stored the data in publicly accessible buckets on AWS cloud storage, did not encrypt it, restrict access to it, log or monitor access to it, or inventory it, and that it was warned at least three times over a two-year period before investigating after a security researcher contacted it in June 2019. - **Access path:** [Public storage, databases and services](/access-paths/public-resource/) - **Kind of control that addresses it:** inventory cloud storage, deny public access to buckets by default, and alert on any bucket or object that becomes publicly readable ## In the words of the source “Vitagene stored in publicly accessible buckets on Amazon Web Service’s (AWS) cloud storage service nearly 2,400 health reports about consumers and raw genetic data of at least 227 consumers sometimes accompanied by a first name” (regulator filing, [FTC press release: FTC Says Genetic Testing Company 1Health Failed to Protect Privacy and Security of DNA Data (web.archive.org copy)](https://web.archive.org/web/2024/https://www.ftc.gov/news-events/news/press-releases/2023/06/ftc-says-genetic-testing-company-1health-failed-protect-privacy-security-dna-data-unfairly-changed)) ## Primary sources - regulator filing: [FTC press release: FTC Says Genetic Testing Company 1Health Failed to Protect Privacy and Security of DNA Data (web.archive.org copy)](https://web.archive.org/web/2024/https://www.ftc.gov/news-events/news/press-releases/2023/06/ftc-says-genetic-testing-company-1health-failed-protect-privacy-security-dna-data-unfairly-changed), dated 16 Jun 2023 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%201Health.io%20Inc.%20(formerly%20Vitagene%2C%20Inc.)%20(vitagene-1health-ftc-2019)) with the entry name in the subject. ## Related pages - [GitHub 2023: rsa ssh key](/incidents/github-rsa-ssh-key-2023/) - [Eatigo International Pte. Ltd. 2023: pdpc](/incidents/eatigo-pdpc-2020/) - [Mailchimp (Intuit) 2023: stolen user credentials and sessi](/incidents/mailchimp-2023/) ## Sources - FTC press release: FTC Says Genetic Testing Company 1Health Failed to Protect Privacy and Security of DNA Data (web.archive.org copy): https://web.archive.org/web/2024/https://www.ftc.gov/news-events/news/press-releases/2023/06/ftc-says-genetic-testing-company-1health-failed-protect-privacy-security-dna-data-unfairly-changed (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/vitagene-1health-ftc-2019/ --- # GitHub (2023): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. GitHub disclosed the incident on 23 Mar 2023. What was exposed, as the sources state it: The company stated GitHub.com's RSA SSH private key was briefly exposed in a public GitHub repository; it stated it had no reason to believe the key was abused and that no GitHub systems or customer information were compromised. - **Disclosed:** 23 Mar 2023 - **Incident window:** not public - **Cloud or provider:** GitHub - **What was exposed:** The company stated GitHub.com's RSA SSH private key was briefly exposed in a public GitHub repository; it stated it had no reason to believe the key was abused and that no GitHub systems or customer information were compromised. - **How access was gained:** The company stated the exposure resulted from what it believes was an inadvertent publishing of private information. - **Access path:** [Leaked access keys and secrets](/access-paths/leaked-access-key/) - **Kind of control that addresses it:** Run secret scanning on all repositories including private-to-public transitions, and keep host private keys in managed key storage ## In the words of the source “This week, we discovered that GitHub.com’s RSA SSH private key was briefly exposed in a public GitHub repository.” (company statement, [We updated our RSA SSH host key](https://github.blog/2023-03-23-we-updated-our-rsa-ssh-host-key/)) ## Primary sources - company statement: [We updated our RSA SSH host key](https://github.blog/2023-03-23-we-updated-our-rsa-ssh-host-key/), dated 23 Mar 2023 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20GitHub%20(github-rsa-ssh-key-2023)) with the entry name in the subject. ## Related pages - [Eatigo International Pte. Ltd. 2023: pdpc](/incidents/eatigo-pdpc-2020/) - [Mailchimp (Intuit) 2023: stolen user credentials and sessi](/incidents/mailchimp-2023/) - [CircleCI 2023: stolen user credentials and session tokens](/incidents/circleci-2023/) ## Sources - We updated our RSA SSH host key: https://github.blog/2023-03-23-we-updated-our-rsa-ssh-host-key/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/github-rsa-ssh-key-2023/ --- # Eatigo International Pte. Ltd. (2023): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Eatigo International Pte. Ltd. disclosed the incident on 10 Mar 2023. What was exposed, as the sources state it: The PDPC decision states a legacy database of approximately 2.76 million users, last updated in late 2018, was accessed and was offered for sale on an online forum. - **Disclosed:** 10 Mar 2023 - **Incident window:** between 2018 and 2020 (exact date not determined) - **Cloud or provider:** not public - **What was exposed:** The PDPC decision states a legacy database of approximately 2.76 million users, last updated in late 2018, was accessed and was offered for sale on an online forum. - **How access was gained:** The decision says the legacy database sat on the infrastructure of a cloud service provider in Singapore, was left out of the company's private network setup, was reachable from the internet to anyone with the credentials, and had no password rotation rule. Staff no longer knew it existed, and no asset inventory or access logs were kept, so the company could not establish how or when access was gained. - **Access path:** [Public storage, databases and services](/access-paths/public-resource/) - **Kind of control that addresses it:** keep an inventory of every cloud database, retire legacy databases, keep databases off the internet, and rotate and monitor database credentials ## In the words of the source “At the time of the Incident, the Affected Database was accessible from the internet and accessible by anyone who had the requisite credentials;” (regulator filing, [PDPC decision: Eatigo International Pte. Ltd.](https://www.pdpc.gov.sg/assets/67f7dfe6-0971-4086-ba23-6239c5fceb26)) ## Primary sources - regulator filing: [PDPC decision: Eatigo International Pte. Ltd.](https://www.pdpc.gov.sg/assets/67f7dfe6-0971-4086-ba23-6239c5fceb26), dated 10 Mar 2023 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Eatigo%20International%20Pte.%20Ltd.%20(eatigo-pdpc-2020)) with the entry name in the subject. ## Related pages - [Mailchimp (Intuit) 2023: stolen user credentials and sessi](/incidents/mailchimp-2023/) - [CircleCI 2023: stolen user credentials and session tokens](/incidents/circleci-2023/) - [Slack 2022: github tokens](/incidents/slack-github-tokens-2022/) ## Sources - PDPC decision: Eatigo International Pte. Ltd.: https://www.pdpc.gov.sg/assets/67f7dfe6-0971-4086-ba23-6239c5fceb26 (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/eatigo-pdpc-2020/ --- # Mailchimp (Intuit) (2023): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Mailchimp (Intuit) disclosed the incident on 13 Jan 2023. What was exposed, as the sources state it: The company stated the incident was limited to 133 Mailchimp accounts and that it found no evidence Intuit systems or customer data beyond those accounts were affected. - **Disclosed:** 13 Jan 2023 - **Incident window:** Identified January 11, 2023; start not public - **Cloud or provider:** Mailchimp - **What was exposed:** The company stated the incident was limited to 133 Mailchimp accounts and that it found no evidence Intuit systems or customer data beyond those accounts were affected. - **How access was gained:** The company stated an unauthorized actor used a social engineering attack on employees and contractors and used the compromised employee credentials to reach an internal customer support and account administration tool. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** Require phishing-resistant MFA for staff and contractors with access to customer-account administration tools and alert on bulk account access ## In the words of the source “The unauthorized actor conducted a social engineering attack on Mailchimp employees and contractors, and obtained access to select Mailchimp accounts using employee credentials compromised in that attack.” (company statement, [Information About a Recent Security Incident](https://mailchimp.com/january-2023-security-incident/)) ## Primary sources - company statement: [Information About a Recent Security Incident](https://mailchimp.com/january-2023-security-incident/), dated 13 Jan 2023 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Mailchimp%20(Intuit)%20(mailchimp-2023)) with the entry name in the subject. ## Related pages - [CircleCI 2023: stolen user credentials and session tokens](/incidents/circleci-2023/) - [Slack 2022: github tokens](/incidents/slack-github-tokens-2022/) - [LastPass 2022: cloud storage](/incidents/lastpass-cloud-storage-2022/) ## Sources - Information About a Recent Security Incident: https://mailchimp.com/january-2023-security-incident/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/mailchimp-2023/ --- # CircleCI (2023): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. CircleCI disclosed the incident on 4 Jan 2023. What was exposed, as the sources state it: The company stated that data from a subset of databases and stores was exfiltrated, including customer environment variables, tokens and keys; the data was encrypted at rest and the third party extracted encryption keys from a running process. - **Disclosed:** 4 Jan 2023 - **Incident window:** Machine compromised December 16, 2022; exfiltration December 22, 2022, as the company states - **Cloud or provider:** CircleCI - **What was exposed:** The company stated that data from a subset of databases and stores was exfiltrated, including customer environment variables, tokens and keys; the data was encrypted at rest and the third party extracted encryption keys from a running process. - **How access was gained:** The company stated that malware on an engineer's laptop stole a valid 2FA-backed SSO session, and that the employee held privileges to generate production access tokens. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/); also [Over-permissioned identities](/access-paths/over-permissioned-identity/) - **Kind of control that addresses it:** Bind sessions to the device and require step-up, phishing-resistant authentication before any identity can generate production access tokens ## In the words of the source “To date, we have learned that an unauthorized third party leveraged malware deployed to a CircleCI engineer’s laptop in order to steal a valid, 2FA-backed SSO session.” (company statement, [Incident report for January 4, 2023 security incident](https://circleci.com/blog/jan-4-2023-incident-report/)) ## Primary sources - company statement: [Incident report for January 4, 2023 security incident](https://circleci.com/blog/jan-4-2023-incident-report/), dated 4 Jan 2023 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20CircleCI%20(circleci-2023)) with the entry name in the subject. ## Related pages - [Slack 2022: github tokens](/incidents/slack-github-tokens-2022/) - [LastPass 2022: cloud storage](/incidents/lastpass-cloud-storage-2022/) - [RedMart Limited 2022: pdpc](/incidents/redmart-pdpc-2020/) ## Sources - Incident report for January 4, 2023 security incident: https://circleci.com/blog/jan-4-2023-incident-report/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/circleci-2023/ --- # Slack (2022): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Slack disclosed the incident on 31 Dec 2022. What was exposed, as the sources state it: The company stated that private code repositories were downloaded, that they contained no customer data, means to access customer data or Slack's primary codebase, and that the production environment was not accessed. - **Disclosed:** 31 Dec 2022 - **Incident window:** December 2022 (repositories downloaded 2022-12-27) - **Cloud or provider:** GitHub - **What was exposed:** The company stated that private code repositories were downloaded, that they contained no customer data, means to access customer data or Slack's primary codebase, and that the production environment was not accessed. - **How access was gained:** The company stated that a limited number of employee tokens were stolen and misused to reach its externally hosted GitHub repository, and that a third-party vendor was compromised. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/); also [Third-party access](/access-paths/third-party-access/) - **Kind of control that addresses it:** store source-control tokens in a managed secret store with short lifetimes and IP allow-listing, and alert on token use from unexpected locations ## In the words of the source “Upon investigation, we discovered that a limited number of Slack employee tokens were stolen and misused to gain access to our externally hosted GitHub repository.” (company statement, [Slack's security update](https://slack.com/blog/news/slack-security-update)) ## Primary sources - company statement: [Slack's security update](https://slack.com/blog/news/slack-security-update), dated 31 Dec 2022 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Slack%20(slack-github-tokens-2022)) with the entry name in the subject. ## Related pages - [LastPass 2022: cloud storage](/incidents/lastpass-cloud-storage-2022/) - [RedMart Limited 2022: pdpc](/incidents/redmart-pdpc-2020/) - [Rackspace Technology, Inc. 2022: hosted exchange](/incidents/rackspace-hosted-exchange-2022/) ## Sources - Slack's security update: https://slack.com/blog/news/slack-security-update (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/slack-github-tokens-2022/ --- # LastPass (2022): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. LastPass disclosed the incident on 22 Dec 2022. What was exposed, as the sources state it: The company stated that a backup copy of customer account information and a backup of customer vault data (with some fields unencrypted, such as website URLs, and others encrypted) were copied from a cloud-based storage service. - **Disclosed:** 22 Dec 2022 - **Incident window:** not public - **Cloud or provider:** not public - **What was exposed:** The company stated that a backup copy of customer account information and a backup of customer vault data (with some fields unencrypted, such as website URLs, and others encrypted) were copied from a cloud-based storage service. - **How access was gained:** The company stated that an unknown threat actor accessed a cloud-based storage environment using information obtained from the August 2022 incident, and that credentials and keys obtained from another targeted employee were used to access and decrypt some storage volumes. - **Access path:** [Leaked access keys and secrets](/access-paths/leaked-access-key/) - **Kind of control that addresses it:** isolate cloud storage access keys and decryption keys from developer and workstation environments, and bind them to short-lived, narrowly scoped identities ## In the words of the source “Based on our investigation to date, we have learned that an unknown threat actor accessed a cloud-based storage environment leveraging information obtained from the incident we previously disclosed in August of 2022.” (company statement, [Notice of Recent Security Incident (archived copy)](https://web.archive.org/web/2023/https://blog.lastpass.com/2022/12/notice-of-recent-security-incident/)) ## Primary sources - company statement: [Notice of Recent Security Incident (archived copy)](https://web.archive.org/web/2023/https://blog.lastpass.com/2022/12/notice-of-recent-security-incident/), dated 22 Dec 2022 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20LastPass%20(lastpass-cloud-storage-2022)) with the entry name in the subject. ## Related pages - [RedMart Limited 2022: pdpc](/incidents/redmart-pdpc-2020/) - [Rackspace Technology, Inc. 2022: hosted exchange](/incidents/rackspace-hosted-exchange-2022/) - [Farrer Park Hospital Pte Ltd 2022: pdpc](/incidents/farrer-park-hospital-pdpc-2018/) ## Sources - Notice of Recent Security Incident (archived copy): https://web.archive.org/web/2023/https://blog.lastpass.com/2022/12/notice-of-recent-security-incident/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/lastpass-cloud-storage-2022/ --- # RedMart Limited (2022): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. RedMart Limited disclosed the incident on 19 Dec 2022. What was exposed, as the sources state it: The PDPC decision states a MongoDB database in the company's Alibaba cloud environment was exfiltrated; its table lists 898,791 affected individuals across customer and seller accounts, with names, emails, contact numbers, addresses, partial credit card details, hashed passwords and call recording links. - **Disclosed:** 19 Dec 2022 - **Incident window:** September 2020; database exfiltrated 6 September 2020 - **Cloud or provider:** AWS and Alibaba Cloud - **What was exposed:** The PDPC decision states a MongoDB database in the company's Alibaba cloud environment was exfiltrated; its table lists 898,791 affected individuals across customer and seller accounts, with names, emails, contact numbers, addresses, partial credit card details, hashed passwords and call recording links. - **How access was gained:** The decision says a threat actor gained unauthorised access to the GitHub account of a member of the DevOps team, found a high-privilege key to the AWS environment in the repositories, used it to find a second key in a private S3 bucket, created an EC2 instance, and crossed a VPN link to the Alibaba cloud database. The database was neither encrypted nor password protected. - **Access path:** [Leaked access keys and secrets](/access-paths/leaked-access-key/); also [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** keep privileged keys out of code repositories, require MFA on every developer account, and require authentication and encryption on databases even behind private links ## In the words of the source “The TA utilised the compromised GitHub user account to search through the GitHub Repositories and found the Chef key.” (regulator filing, [PDPC decision: RedMart Limited [2022] SGPDPC 8](https://www.pdpc.gov.sg/assets/438fd68d-bcd9-4d25-8f77-09268e80a4e4)) ## Primary sources - regulator filing: [PDPC decision: RedMart Limited [2022] SGPDPC 8](https://www.pdpc.gov.sg/assets/438fd68d-bcd9-4d25-8f77-09268e80a4e4), dated 19 Dec 2022 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20RedMart%20Limited%20(redmart-pdpc-2020)) with the entry name in the subject. ## Related pages - [Rackspace Technology, Inc. 2022: hosted exchange](/incidents/rackspace-hosted-exchange-2022/) - [Farrer Park Hospital Pte Ltd 2022: pdpc](/incidents/farrer-park-hospital-pdpc-2018/) - [Dropbox 2022: github phishing](/incidents/dropbox-github-phishing-2022/) ## Sources - PDPC decision: RedMart Limited [2022] SGPDPC 8: https://www.pdpc.gov.sg/assets/438fd68d-bcd9-4d25-8f77-09268e80a4e4 (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/redmart-pdpc-2020/ --- # Rackspace Technology, Inc. (2022): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Rackspace Technology, Inc. disclosed the incident on 6 Dec 2022. The sources do not state how access was gained, so the access path reads not public. The cloud or provider involved: Rackspace Hosted Exchange. Every field rests on the primary sources listed below. - **Disclosed:** 6 Dec 2022 - **Incident window:** Not stated; the December 9, 2022 update refers to the recent ransomware incident - **Cloud or provider:** Rackspace Hosted Exchange - **What was exposed:** The filing says a ransomware incident affected the company's Hosted Exchange email business, a managed email solution for small and medium businesses, and that the incident was limited solely to that product (about 1 percent of annual revenue). It says affected customers were being moved to Microsoft Office 365. What data was taken is not public. - **How access was gained:** The filing does not say how access was gained. How access was gained is not public. - **Access path:** Not public: the sources do not state how access was gained - **Kind of control that addresses it:** not public (the sources do not state how access was gained) ## In the words of the source “On December 9, 2022, Rackspace Technology, Inc. (the Company) issued an update following the recent ransomware incident affecting its Hosted Exchange email business, a managed email solution provided to small and medium businesses.” (regulator filing, [Rackspace Technology Form 8-K, Item 7.01, filed 2022-12-09](https://www.sec.gov/Archives/edgar/data/1810019/000119312522301429/d429186d8k.htm)) ## Primary sources - regulator filing: [Rackspace Technology Form 8-K, Item 7.01, filed 2022-12-09](https://www.sec.gov/Archives/edgar/data/1810019/000119312522301429/d429186d8k.htm), dated 9 Dec 2022 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Rackspace%20Technology%2C%20Inc.%20(rackspace-hosted-exchange-2022)) with the entry name in the subject. ## Related pages - [Farrer Park Hospital Pte Ltd 2022: pdpc](/incidents/farrer-park-hospital-pdpc-2018/) - [Dropbox 2022: github phishing](/incidents/dropbox-github-phishing-2022/) - [Chegg 2022: ftc](/incidents/chegg-ftc-2022/) ## Sources - Rackspace Technology Form 8-K, Item 7.01, filed 2022-12-09: https://www.sec.gov/Archives/edgar/data/1810019/000119312522301429/d429186d8k.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/rackspace-hosted-exchange-2022/ --- # Farrer Park Hospital Pte Ltd (2022): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Farrer Park Hospital Pte Ltd disclosed the incident on 18 Nov 2022. What was exposed, as the sources state it: The PDPC decision states 9,271 emails were automatically forwarded from two employees' Microsoft Office 365 mailboxes to a third-party address, disclosing the personal data of 3,539 individuals, 1,923 of them with medical information. - **Disclosed:** 18 Nov 2022 - **Incident window:** 8 March 2018 to 25 October 2019 - **Cloud or provider:** Microsoft 365 - **What was exposed:** The PDPC decision states 9,271 emails were automatically forwarded from two employees' Microsoft Office 365 mailboxes to a third-party address, disclosing the personal data of 3,539 individuals, 1,923 of them with medical information. - **How access was gained:** The decision says the two mailboxes had been configured to forward all incoming emails to a third party. It was found on 24 October 2019 when Office 365 restricted the accounts, which the decision describes as a security feature indicating unauthorised access. The decision does not state how the forwarding was set up; multi-factor authentication was not in place when it began in March 2018. - **Access path:** Not public: the sources do not state how access was gained - **Kind of control that addresses it:** disable automatic external mail forwarding by default, require MFA for all mailboxes, and alert on new forwarding rules ## In the words of the source “Further investigations by the Organisation confirmed that the Email Accounts had been configured to automatically forward all incoming emails to the Third Party.” (regulator filing, [PDPC decision: Farrer Park Hospital Pte Ltd [2022]](https://www.pdpc.gov.sg/assets/685c6f15-4fd9-49be-88e4-b86975ce7974)) ## Primary sources - regulator filing: [PDPC decision: Farrer Park Hospital Pte Ltd [2022]](https://www.pdpc.gov.sg/assets/685c6f15-4fd9-49be-88e4-b86975ce7974), dated 18 Nov 2022 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Farrer%20Park%20Hospital%20Pte%20Ltd%20(farrer-park-hospital-pdpc-2018)) with the entry name in the subject. ## Related pages - [Dropbox 2022: github phishing](/incidents/dropbox-github-phishing-2022/) - [Chegg 2022: ftc](/incidents/chegg-ftc-2022/) - [Drizly 2022: ftc](/incidents/drizly-ftc-2022/) ## Sources - PDPC decision: Farrer Park Hospital Pte Ltd [2022]: https://www.pdpc.gov.sg/assets/685c6f15-4fd9-49be-88e4-b86975ce7974 (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/farrer-park-hospital-pdpc-2018/ --- # Dropbox (2022): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Dropbox disclosed the incident on 1 Nov 2022. What was exposed, as the sources state it: The company stated that the threat actor copied 130 code repositories containing some credentials (primarily API keys) and a few thousand names and email addresses; no customer content, passwords or payment information was accessed. - **Disclosed:** 1 Nov 2022 - **Incident window:** October 2022 (GitHub alert 2022-10-14, activity began the previous day) - **Cloud or provider:** GitHub - **What was exposed:** The company stated that the threat actor copied 130 code repositories containing some credentials (primarily API keys) and a few thousand names and email addresses; no customer content, passwords or payment information was accessed. - **How access was gained:** The company stated that employees received phishing emails impersonating a CI service, and that a fake login page collected GitHub credentials and a one-time code, giving the threat actor access to one GitHub organization. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/); also [Leaked access keys and secrets](/access-paths/leaked-access-key/) - **Kind of control that addresses it:** require phishing-resistant authentication (WebAuthn) for source-control accounts and keep API keys out of repositories ## In the words of the source “These legitimate-looking emails directed employees to visit a fake CircleCI login page, enter their GitHub username and password, and then use their hardware authentication key to pass a One Time Password (OTP) to the malicious site.” (company statement, [A recent phishing campaign targeting Dropbox](https://dropbox.tech/security/a-recent-phishing-campaign-targeting-dropbox)) ## Primary sources - company statement: [A recent phishing campaign targeting Dropbox](https://dropbox.tech/security/a-recent-phishing-campaign-targeting-dropbox), dated 1 Nov 2022 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Dropbox%20(dropbox-github-phishing-2022)) with the entry name in the subject. ## Related pages - [Chegg 2022: ftc](/incidents/chegg-ftc-2022/) - [Drizly 2022: ftc](/incidents/drizly-ftc-2022/) - [Microsoft (misconfigured storage endpoint) 2022: public st](/incidents/microsoft-misconfigured-endpoint-2022/) ## Sources - A recent phishing campaign targeting Dropbox: https://dropbox.tech/security/a-recent-phishing-campaign-targeting-dropbox (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/dropbox-github-phishing-2022/ --- # Chegg (2022): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Chegg disclosed the incident on 31 Oct 2022. What was exposed, as the sources state it: The FTC stated that a third-party cloud database held personal information of approximately 40 million customers, including names, email addresses and passwords and, for certain users, scholarship data. - **Disclosed:** 31 Oct 2022 - **Incident window:** 2017 to 2020 (four breaches per the FTC; the cloud database breach followed the September 2017 incident by less than a year) - **Cloud or provider:** not public - **What was exposed:** The FTC stated that a third-party cloud database held personal information of approximately 40 million customers, including names, email addresses and passwords and, for certain users, scholarship data. - **How access was gained:** The FTC stated that a former contractor used login information the company shared with employees and outside contractors to access one of Chegg's third-party cloud databases. - **Access path:** [Retained access after departure or change](/access-paths/retained-access/) - **Kind of control that addresses it:** issue individual, revocable credentials to contractors, remove access at offboarding, and require MFA for access to cloud databases ## In the words of the source “a former Chegg contractor used login information the company shared with employees and outside contractors to access one of Chegg’s third-party cloud databases containing personal information of approximately 40 million customers.” (regulator filing, [FTC Brings Action Against Ed Tech Provider Chegg (archived copy)](https://web.archive.org/web/2023/https://www.ftc.gov/news-events/news/press-releases/2022/10/ftc-brings-action-against-ed-tech-provider-chegg-careless-security-exposed-personal-data-millions)) ## Primary sources - regulator filing: [FTC Brings Action Against Ed Tech Provider Chegg (archived copy)](https://web.archive.org/web/2023/https://www.ftc.gov/news-events/news/press-releases/2022/10/ftc-brings-action-against-ed-tech-provider-chegg-careless-security-exposed-personal-data-millions), dated 31 Oct 2022 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Chegg%20(chegg-ftc-2022)) with the entry name in the subject. ## Related pages - [Drizly 2022: ftc](/incidents/drizly-ftc-2022/) - [Microsoft (misconfigured storage endpoint) 2022: public st](/incidents/microsoft-misconfigured-endpoint-2022/) - [Uber 2022: stolen user credentials and session tokens](/incidents/uber-2022/) ## Sources - FTC Brings Action Against Ed Tech Provider Chegg (archived copy): https://web.archive.org/web/2023/https://www.ftc.gov/news-events/news/press-releases/2022/10/ftc-brings-action-against-ed-tech-provider-chegg-careless-security-exposed-personal-data-millions (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/chegg-ftc-2022/ --- # Drizly (2022): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Drizly disclosed the incident on 24 Oct 2022. What was exposed, as the sources state it: The FTC complaint stated that Drizly's databases held personal information of more than 2.5 million consumers, that the intruder took a user table of more than 2.5 million records, and that the company hosted its production database on Amazon RDS, an AWS service. - **Disclosed:** 24 Oct 2022 - **Incident window:** April 2018 (executive granted GitHub repository access) and early July 2020 (intrusion) - **Cloud or provider:** AWS - **What was exposed:** The FTC complaint stated that Drizly's databases held personal information of more than 2.5 million consumers, that the intruder took a user table of more than 2.5 million records, and that the company hosted its production database on Amazon RDS, an AWS service. - **How access was gained:** The FTC complaint stated that in early July 2020 a malicious actor reused credentials from an unrelated breach to reach an executive's GitHub account, found AWS and database credentials stored in the company's GitHub repositories, and used them to change the company's AWS security settings and reach the production environment. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/); also [Leaked access keys and secrets](/access-paths/leaked-access-key/) - **Kind of control that addresses it:** require MFA on source-control accounts, never store cloud credentials in repositories, and limit employee access to personal data ## In the words of the source “The intruder used the compromised credentials from Drizly’s GitHub repositories to modify the company’s AWS security settings.” (regulator filing, [FTC complaint, Drizly LLC](https://www.ftc.gov/system/files/ftc_gov/pdf/202-3185-Drizly-Complaint.pdf)) ## Primary sources - regulator filing: [FTC complaint, Drizly LLC](https://www.ftc.gov/system/files/ftc_gov/pdf/202-3185-Drizly-Complaint.pdf), dated 24 Oct 2022 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Drizly%20(drizly-ftc-2022)) with the entry name in the subject. ## Related pages - [Microsoft (misconfigured storage endpoint) 2022: public st](/incidents/microsoft-misconfigured-endpoint-2022/) - [Uber 2022: stolen user credentials and session tokens](/incidents/uber-2022/) - [GitHub users (phishing campaign reported by GitHub) 2022: ](/incidents/github-circleci-phishing-2022/) ## Sources - FTC complaint, Drizly LLC: https://www.ftc.gov/system/files/ftc_gov/pdf/202-3185-Drizly-Complaint.pdf (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/drizly-ftc-2022/ --- # Microsoft (misconfigured storage endpoint) (2022): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Microsoft (misconfigured storage endpoint) disclosed the incident on 19 Oct 2022. The ledger reads the access path as public storage, databases and services. The cloud or provider involved: not public. Every field rests on the primary sources listed below. Fields no primary source states read not public. - **Disclosed:** 19 Oct 2022 - **Incident window:** not public - **Cloud or provider:** not public - **What was exposed:** The post-mortem says business transaction data (names, email addresses, email content, company names, phone numbers and possibly attached files) from interactions between Microsoft and prospective customers could be reached without authentication; it says the scope reported by the researchers was exaggerated and does not give its own count. - **How access was gained:** The post-mortem says an unintentional misconfiguration on an endpoint, not a security vulnerability, allowed unauthenticated access; the endpoint is now accessible only with authentication. - **Access path:** [Public storage, databases and services](/access-paths/public-resource/) - **Kind of control that addresses it:** deny anonymous access to storage endpoints by default and continuously scan for publicly reachable storage ## In the words of the source “The issue was caused by an unintentional misconfiguration on an endpoint that is not in use across the Microsoft ecosystem and was not the result of a security vulnerability.” (provider post mortem, [Investigation Regarding Misconfigured Microsoft Storage Location (archived copy)](https://web.archive.org/web/2022/https://msrc-blog.microsoft.com/2022/10/19/investigation-regarding-misconfigured-microsoft-storage-location-2/)) ## Primary sources - provider post mortem: [Investigation Regarding Misconfigured Microsoft Storage Location (archived copy)](https://web.archive.org/web/2022/https://msrc-blog.microsoft.com/2022/10/19/investigation-regarding-misconfigured-microsoft-storage-location-2/), dated 19 Oct 2022 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Microsoft%20(misconfigured%20storage%20endpoint)%20(microsoft-misconfigured-endpoint-2022)) with the entry name in the subject. ## Related pages - [Uber 2022: stolen user credentials and session tokens](/incidents/uber-2022/) - [GitHub users (phishing campaign reported by GitHub) 2022: ](/incidents/github-circleci-phishing-2022/) - [MyRepublic Limited 2022: pdpc](/incidents/myrepublic-pdpc-2021/) ## Sources - Investigation Regarding Misconfigured Microsoft Storage Location (archived copy): https://web.archive.org/web/2022/https://msrc-blog.microsoft.com/2022/10/19/investigation-regarding-misconfigured-microsoft-storage-location-2/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/microsoft-misconfigured-endpoint-2022/ --- # Uber (2022): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Uber disclosed the incident on 16 Sep 2022. What was exposed, as the sources state it: The company stated that the attacker accessed several internal systems, downloaded some internal Slack messages, accessed information from an internal finance tool and its HackerOne dashboard, and that it had not found access to customer or user data held by its cloud providers. - **Disclosed:** 16 Sep 2022 - **Incident window:** September 2022 - **Cloud or provider:** Google Workspace - **What was exposed:** The company stated that the attacker accessed several internal systems, downloaded some internal Slack messages, accessed information from an internal finance tool and its HackerOne dashboard, and that it had not found access to customer or user data held by its cloud providers. - **How access was gained:** The company stated that an attacker likely bought a contractor's corporate password after malware on the contractor's personal device exposed it, then repeated two-factor login requests until the contractor accepted one. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/); also [Over-permissioned identities](/access-paths/over-permissioned-identity/) - **Kind of control that addresses it:** require phishing-resistant MFA that cannot be satisfied by an approval prompt, and rate-limit and alert on repeated MFA requests ## In the words of the source “It is likely that the attacker purchased the contractor’s Uber corporate password on the dark web, after the contractor’s personal device had been infected with malware, exposing those credentials.” (company statement, [Security update](https://www.uber.com/newsroom/security-update/)) ## Primary sources - company statement: [Security update](https://www.uber.com/newsroom/security-update/), dated 19 Sep 2022 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Uber%20(uber-2022)) with the entry name in the subject. ## Related pages - [GitHub users (phishing campaign reported by GitHub) 2022: ](/incidents/github-circleci-phishing-2022/) - [MyRepublic Limited 2022: pdpc](/incidents/myrepublic-pdpc-2021/) - [Mailchimp (Intuit) 2022: crypto social engineering](/incidents/mailchimp-crypto-social-engineering-2022/) ## Sources - Security update: https://www.uber.com/newsroom/security-update/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/uber-2022/ --- # GitHub users (phishing campaign reported by GitHub) (2022): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. GitHub users (phishing campaign reported by GitHub) disclosed the incident on 16 Sep 2022. What was exposed, as the sources state it: GitHub stated that the campaign affected many victim organizations, and that in many cases the threat actor downloaded private repository contents accessible to the compromised user; GitHub itself was not affected. - **Disclosed:** 16 Sep 2022 - **Incident window:** From September 2022 - **Cloud or provider:** GitHub - **What was exposed:** GitHub stated that the campaign affected many victim organizations, and that in many cases the threat actor downloaded private repository contents accessible to the compromised user; GitHub itself was not affected. - **How access was gained:** GitHub stated that a phishing site imitating the GitHub login page stole credentials and relayed TOTP codes to the threat actor in real time, and that accounts protected by hardware security keys were not vulnerable. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** require hardware security keys or other phishing-resistant authentication for source-control accounts ## In the words of the source “Clicking the link takes the user to a phishing site that looks like the GitHub login page but steals any credentials entered.” (company statement, [Security alert: new phishing campaign targets GitHub users](https://github.blog/2022-09-16-security-alert-new-phishing-campaign-targets-github-users/)) ## Primary sources - company statement: [Security alert: new phishing campaign targets GitHub users](https://github.blog/2022-09-16-security-alert-new-phishing-campaign-targets-github-users/), dated 16 Sep 2022 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20GitHub%20users%20(phishing%20campaign%20reported%20by%20GitHub)%20(github-circleci-phishing-2022)) with the entry name in the subject. ## Related pages - [MyRepublic Limited 2022: pdpc](/incidents/myrepublic-pdpc-2021/) - [Mailchimp (Intuit) 2022: crypto social engineering](/incidents/mailchimp-crypto-social-engineering-2022/) - [Cisco Systems 2022: google account vishing](/incidents/cisco-google-account-vishing-2022/) ## Sources - Security alert: new phishing campaign targets GitHub users: https://github.blog/2022-09-16-security-alert-new-phishing-campaign-targets-github-users/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/github-circleci-phishing-2022/ --- # MyRepublic Limited (2022): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. MyRepublic Limited disclosed the incident on 15 Sep 2022. What was exposed, as the sources state it: The PDPC decision states the personal data of 79,388 customers was accessed and exfiltrated: scanned copies of identity cards and work passes submitted as KYC documents and number portability documents. - **Disclosed:** 15 Sep 2022 - **Incident window:** 29 August 2021 (discovered) - **Cloud or provider:** AWS - **What was exposed:** The PDPC decision states the personal data of 79,388 customers was accessed and exfiltrated: scanned copies of identity cards and work passes submitted as KYC documents and number portability documents. - **How access was gained:** The decision says the Mobile Order Portal stored KYC documents in a bucket on AWS cloud storage, and the bucket's access key sat in the portal's source code. The external actor used the key to access the bucket. The company could not determine with certainty how the key was obtained but found it was disclosed in a portal page showing technical information and in a source repository open to all its developers. - **Access path:** [Leaked access keys and secrets](/access-paths/leaked-access-key/) - **Kind of control that addresses it:** keep storage keys out of source code and technical debug pages, restrict repository access to those who need it, and use short-lived scoped credentials for storage access ## In the words of the source “Investigations revealed that the external actor had utilised the Access Key to access the Bucket.” (regulator filing, [PDPC decision: MyRepublic Limited [2022] SGPDPC 5](https://www.pdpc.gov.sg/assets/652104b6-c8a1-4a7e-bc7c-2e1b4d87679b)) ## Primary sources - regulator filing: [PDPC decision: MyRepublic Limited [2022] SGPDPC 5](https://www.pdpc.gov.sg/assets/652104b6-c8a1-4a7e-bc7c-2e1b4d87679b), dated 15 Sep 2022 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20MyRepublic%20Limited%20(myrepublic-pdpc-2021)) with the entry name in the subject. ## Related pages - [Mailchimp (Intuit) 2022: crypto social engineering](/incidents/mailchimp-crypto-social-engineering-2022/) - [Cisco Systems 2022: google account vishing](/incidents/cisco-google-account-vishing-2022/) - [Cloudflare 2022: phishing](/incidents/cloudflare-phishing-2022/) ## Sources - PDPC decision: MyRepublic Limited [2022] SGPDPC 5: https://www.pdpc.gov.sg/assets/652104b6-c8a1-4a7e-bc7c-2e1b4d87679b (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/myrepublic-pdpc-2021/ --- # Mailchimp (Intuit) (2022): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Mailchimp (Intuit) disclosed the incident on 12 Aug 2022. What was exposed, as the sources state it: The company stated 214 Mailchimp accounts were affected, in a targeted incident focused on users in industries related to cryptocurrency and finance. The ledger reads the access path as stolen user credentials and session tokens with over-permissioned identities. - **Disclosed:** 12 Aug 2022 - **Incident window:** August 8 to August 2022 (the company) - **Cloud or provider:** Mailchimp - **What was exposed:** The company stated 214 Mailchimp accounts were affected, in a targeted incident focused on users in industries related to cryptocurrency and finance. - **How access was gained:** The company stated an unauthorized actor accessed an internal tool used by customer-facing teams for customer support and account administration, after a social engineering attack on Mailchimp employees that obtained employee credentials. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/); also [Over-permissioned identities](/access-paths/over-permissioned-identity/) - **Kind of control that addresses it:** require phishing-resistant MFA for support staff and limit what a support tool can do to customer accounts ## In the words of the source “The incident was propagated by an unauthorized actor who conducted a social engineering attack on Mailchimp employees, and obtained access using employee credentials compromised in that social engineering attack.” (company statement, [Information About a Recent Security Incident Targeting Crypto Companies (Mailchimp)](https://mailchimp.com/august-2022-security-incident/)) ## Primary sources - company statement: [Information About a Recent Security Incident Targeting Crypto Companies (Mailchimp)](https://mailchimp.com/august-2022-security-incident/), dated 12 Aug 2022 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Mailchimp%20(Intuit)%20(mailchimp-crypto-social-engineering-2022)) with the entry name in the subject. ## Related pages - [Cisco Systems 2022: google account vishing](/incidents/cisco-google-account-vishing-2022/) - [Cloudflare 2022: phishing](/incidents/cloudflare-phishing-2022/) - [Twilio 2022: stolen user credentials and session tokens](/incidents/twilio-2022/) ## Sources - Information About a Recent Security Incident Targeting Crypto Companies (Mailchimp): https://mailchimp.com/august-2022-security-incident/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/mailchimp-crypto-social-engineering-2022/ --- # Cisco Systems (2022): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Cisco Systems disclosed the incident on 10 Aug 2022. What was exposed, as the sources state it: The company stated the only successful data exfiltration was the contents of a Box folder associated with a compromised employee's account and employee authentication data from Active Directory, and that the Box data was not sensitive. - **Disclosed:** 10 Aug 2022 - **Incident window:** Cisco became aware of a potential compromise on May 24, 2022 - **Cloud or provider:** Box - **What was exposed:** The company stated the only successful data exfiltration was the contents of a Box folder associated with a compromised employee's account and employee authentication data from Active Directory, and that the Box data was not sensitive. - **How access was gained:** The company stated an employee's credentials were compromised after an attacker gained control of a personal Google account to which credentials saved in the browser were synchronised, and the attacker then used voice phishing and repeated multi-factor push requests until one was accepted, which gave access to the VPN. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** require phishing-resistant multi-factor authentication, block saving of work credentials in personal browser profiles, and alert on new device enrollment for MFA ## In the words of the source “During the investigation, it was determined that a Cisco employee’s credentials were compromised after an attacker gained control of a personal Google account where credentials saved in the victim’s browser were being synchronized.” (company statement, [Cisco Talos shares insights related to recent cyber attack on Cisco](https://blog.talosintelligence.com/recent-cyber-attack/)) ## Primary sources - company statement: [Cisco Talos shares insights related to recent cyber attack on Cisco](https://blog.talosintelligence.com/recent-cyber-attack/), dated 10 Aug 2022 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Cisco%20Systems%20(cisco-google-account-vishing-2022)) with the entry name in the subject. ## Related pages - [Cloudflare 2022: phishing](/incidents/cloudflare-phishing-2022/) - [Twilio 2022: stolen user credentials and session tokens](/incidents/twilio-2022/) - [Crawfort Pte. Ltd. 2022: pdpc](/incidents/crawfort-pdpc-2020/) ## Sources - Cisco Talos shares insights related to recent cyber attack on Cisco: https://blog.talosintelligence.com/recent-cyber-attack/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/cisco-google-account-vishing-2022/ --- # Cloudflare (2022): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Cloudflare disclosed the incident on 9 Aug 2022. What was exposed, as the sources state it: The company stated that no Cloudflare systems were compromised. The ledger reads the access path as stolen user credentials and session tokens. The cloud or provider involved: Okta. - **Disclosed:** 9 Aug 2022 - **Incident window:** From 2022-07-20 (phishing messages began that day) - **Cloud or provider:** Okta - **What was exposed:** The company stated that no Cloudflare systems were compromised. - **How access was gained:** The company stated that employees received text messages pointing to a look-alike page for its identity provider, which asked for a username, password and a one-time code; individual employees did enter credentials, but physical security keys required for its applications stopped the attacker. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** issue physical security keys to every employee and require them for all applications so that relayed passwords and one-time codes cannot be used ## In the words of the source “The phishing page prompted anyone who visited it for their username and password.” (company statement, [The mechanics of a sophisticated phishing scam and how we stopped it](https://blog.cloudflare.com/2022-07-sms-phishing-attacks/)) ## Primary sources - company statement: [The mechanics of a sophisticated phishing scam and how we stopped it](https://blog.cloudflare.com/2022-07-sms-phishing-attacks/), dated 9 Aug 2022 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Cloudflare%20(cloudflare-phishing-2022)) with the entry name in the subject. ## Related pages - [Twilio 2022: stolen user credentials and session tokens](/incidents/twilio-2022/) - [Crawfort Pte. Ltd. 2022: pdpc](/incidents/crawfort-pdpc-2020/) - [Quoine Pte Ltd 2022: pdpc](/incidents/quoine-pdpc-2020/) ## Sources - The mechanics of a sophisticated phishing scam and how we stopped it: https://blog.cloudflare.com/2022-07-sms-phishing-attacks/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/cloudflare-phishing-2022/ --- # Twilio (2022): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Twilio disclosed the incident on 7 Aug 2022. What was exposed, as the sources state it: The company stated that 163 Twilio customers, out of a customer base of over 270,000, had data accessed without authorization, and that 93 individual Authy users, out of approximately 75 million, had additional devices registered to their accounts. - **Disclosed:** 7 Aug 2022 - **Incident window:** August 2022 (company became aware 2022-08-04) - **Cloud or provider:** Twilio - **What was exposed:** The company stated that 163 Twilio customers, out of a customer base of over 270,000, had data accessed without authorization, and that 93 individual Authy users, out of approximately 75 million, had additional devices registered to their accounts. - **How access was gained:** The company stated that text messages sent to current and former employees led some of them to a page impersonating the Twilio sign-in page, and that the attackers used the stolen credentials to reach internal systems. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** require phishing-resistant MFA (hardware-bound credentials) for every workforce identity with access to customer data systems ## In the words of the source “The attackers then used the stolen credentials to gain access to some of our internal systems, where they were able to access certain customer data.” (company statement, [Incident Report: Employee and Customer Account Compromise](https://www.twilio.com/blog/august-2022-social-engineering-attack)) ## Primary sources - company statement: [Incident Report: Employee and Customer Account Compromise](https://www.twilio.com/blog/august-2022-social-engineering-attack), dated 7 Aug 2022 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Twilio%20(twilio-2022)) with the entry name in the subject. ## Related pages - [Crawfort Pte. Ltd. 2022: pdpc](/incidents/crawfort-pdpc-2020/) - [Quoine Pte Ltd 2022: pdpc](/incidents/quoine-pdpc-2020/) - [Wegmans 2022: azure containers](/incidents/wegmans-azure-containers-2021/) ## Sources - Incident Report: Employee and Customer Account Compromise: https://www.twilio.com/blog/august-2022-social-engineering-attack (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/twilio-2022/ --- # Crawfort Pte. Ltd. (2022): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Crawfort Pte. Ltd. disclosed the incident on 14 Jul 2022. What was exposed, as the sources state it: The PDPC decision states the personal data of 5,421 customers was affected: images of identity cards, PDF loan contracts and income documents. - **Disclosed:** 14 Jul 2022 - **Incident window:** about 15 April 2020 (one-week migration window); dark web sale notified 9 June 2021 - **Cloud or provider:** AWS - **What was exposed:** The PDPC decision states the personal data of 5,421 customers was affected: images of identity cards, PDF loan contracts and income documents. - **How access was gained:** The decision says the company opened an S3 server port in its AWS environment for one week during a data migration, and that the S3 bucket holding the data was publicly accessible due to a misconfiguration. It states the threat actor gained access to the public bucket in that period. - **Access path:** [Public storage, databases and services](/access-paths/public-resource/); also [Exposed network services and request paths](/access-paths/exposed-network-service/) - **Kind of control that addresses it:** require a risk review before temporary changes to cloud storage, block public access on buckets, and alert on any opened ports or public buckets ## In the words of the source “The investigation identified an opened S3 server port in the Organisation’s AWS environment as the cause of the Incident.” (regulator filing, [PDPC decision: Crawfort Pte. Ltd. (7 June 2022)](https://www.pdpc.gov.sg/assets/e476cec1-2618-4d2d-b837-99109a8a97f4)) ## Primary sources - regulator filing: [PDPC decision: Crawfort Pte. Ltd. (7 June 2022)](https://www.pdpc.gov.sg/assets/e476cec1-2618-4d2d-b837-99109a8a97f4), dated 14 Jul 2022 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Crawfort%20Pte.%20Ltd.%20(crawfort-pdpc-2020)) with the entry name in the subject. ## Related pages - [Quoine Pte Ltd 2022: pdpc](/incidents/quoine-pdpc-2020/) - [Wegmans 2022: azure containers](/incidents/wegmans-azure-containers-2021/) - [Wegmans Food Markets, Inc. 2022: azure container](/incidents/wegmans-azure-container-2018/) ## Sources - PDPC decision: Crawfort Pte. Ltd. (7 June 2022): https://www.pdpc.gov.sg/assets/e476cec1-2618-4d2d-b837-99109a8a97f4 (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/crawfort-pdpc-2020/ --- # Quoine Pte Ltd (2022): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Quoine Pte Ltd disclosed the incident on 14 Jul 2022. What was exposed, as the sources state it: The PDPC decision states the personal data of 652,564 customers was accessed and exfiltrated, including names, addresses, emails, identity document images for 362,035 customers, transaction information and bank account details. - **Disclosed:** 14 Jul 2022 - **Incident window:** 13 to 17 November 2020 - **Cloud or provider:** not public - **What was exposed:** The PDPC decision states the personal data of 652,564 customers was accessed and exfiltrated, including names, addresses, emails, identity document images for 362,035 customers, transaction information and bank account details. - **How access was gained:** The decision says social engineering of employees of the company's domain registrar led to control of its domain hosting account being handed to an external actor, who redirected the company's mail, reset the password of a DevOps account, and used the reset credentials to reach the company's cloud computing platform, which held API keys to its databases. - **Access path:** [Third-party access](/access-paths/third-party-access/); also [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** require strong identity checks and MFA on domain and registrar accounts, keep API keys out of reach of a single automation account, and monitor mail routing changes ## In the words of the source “As a result of social engineering attacks on employees of the Domain Provider, an employee of the Domain Provider incorrectly transferred control of the Organisation’s Domain Hosting Account to an external actor.” (regulator filing, [PDPC decision: Quoine Pte Ltd [2022] SGPDPC 2](https://www.pdpc.gov.sg/assets/5ac38f55-f48d-47af-9988-b926bb999b03)) ## Primary sources - regulator filing: [PDPC decision: Quoine Pte Ltd [2022] SGPDPC 2](https://www.pdpc.gov.sg/assets/5ac38f55-f48d-47af-9988-b926bb999b03), dated 14 Jul 2022 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Quoine%20Pte%20Ltd%20(quoine-pdpc-2020)) with the entry name in the subject. ## Related pages - [Wegmans 2022: azure containers](/incidents/wegmans-azure-containers-2021/) - [Wegmans Food Markets, Inc. 2022: azure container](/incidents/wegmans-azure-container-2018/) - [MGM Resorts International 2022: opc](/incidents/mgm-resorts-opc-2019/) ## Sources - PDPC decision: Quoine Pte Ltd [2022] SGPDPC 2: https://www.pdpc.gov.sg/assets/5ac38f55-f48d-47af-9988-b926bb999b03 (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/quoine-pdpc-2020/ --- # Wegmans (2022): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Wegmans disclosed the incident on 30 Jun 2022. The ledger reads the access path as public storage, databases and services. The cloud or provider involved: Microsoft Azure. Every field rests on the primary sources listed below. Fields no primary source states read not public. - **Disclosed:** 30 Jun 2022 - **Incident window:** container misconfigured from January 2018 to April 2021; a second from November 2018, as the release says - **Cloud or provider:** Microsoft Azure - **What was exposed:** The release says a database backup with over three million records of customer email addresses and account passwords sat in the first container; a second container held a database with customers’ names, email addresses and mailing addresses; the release puts the total at more than three million consumers, including more than 830,000 New Yorkers. - **How access was gained:** A security researcher told the company in April 2021 that a cloud storage container hosted on Microsoft Azure was open to public access; the release says a second container had been public since November 2018. - **Access path:** [Public storage, databases and services](/access-paths/public-resource/) - **Kind of control that addresses it:** block public access at the storage-account level, deny it by policy and alert on any container that becomes reachable without authentication ## In the words of the source “a security researcher informed Wegmans that a cloud storage container hosted on Microsoft Azure was left unsecured and open to public access, potentially exposing consumers’ sensitive information.” (regulator filing, [NY Attorney General: Attorney General James Secures $400,000 From Wegmans After Data Breach Exposed Consumers’ Personal Information](https://ag.ny.gov/press-release/2022/attorney-general-james-secures-400000-wegmans-after-data-breach-exposed-consumers)) ## Primary sources - regulator filing: [NY Attorney General: Attorney General James Secures $400,000 From Wegmans After Data Breach Exposed Consumers’ Personal Information](https://ag.ny.gov/press-release/2022/attorney-general-james-secures-400000-wegmans-after-data-breach-exposed-consumers), dated 30 Jun 2022 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Wegmans%20(wegmans-azure-containers-2021)) with the entry name in the subject. ## Related pages - [Wegmans Food Markets, Inc. 2022: azure container](/incidents/wegmans-azure-container-2018/) - [MGM Resorts International 2022: opc](/incidents/mgm-resorts-opc-2019/) - [GeniusU Pte. Ltd. 2022: pdpc](/incidents/geniusu-pdpc-2021/) ## Sources - NY Attorney General: Attorney General James Secures $400,000 From Wegmans After Data Breach Exposed Consumers’ Personal Information: https://ag.ny.gov/press-release/2022/attorney-general-james-secures-400000-wegmans-after-data-breach-exposed-consumers (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/wegmans-azure-containers-2021/ --- # Wegmans Food Markets, Inc. (2022): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Wegmans Food Markets, Inc. disclosed the incident on 30 Jun 2022. What was exposed, as the sources state it: The New York Attorney General states the data of more than three million consumers nationwide, including more than 830,000 New Yorkers, was exposed, including a database backup file with over three million records of customer email addresses and account passwords. - **Disclosed:** 30 Jun 2022 - **Incident window:** January 2018 to April 2021 (the container was misconfigured from its creation in January 2018 until April 2021, per the New York Attorney General) - **Cloud or provider:** Microsoft Azure - **What was exposed:** The New York Attorney General states the data of more than three million consumers nationwide, including more than 830,000 New Yorkers, was exposed, including a database backup file with over three million records of customer email addresses and account passwords. - **How access was gained:** The New York Attorney General states that in April 2021 a security researcher informed the company that a cloud storage container hosted on Microsoft Azure was left unsecured and open to public access. The Attorney General’s office found the company failed to configure the container to limit access, to inventory its cloud assets, to secure all user passwords and to regularly conduct security testing of its cloud assets. - **Access path:** [Public storage, databases and services](/access-paths/public-resource/) - **Kind of control that addresses it:** keep an inventory of cloud storage, deny anonymous access to storage containers by default, and continuously test storage configuration for public exposure ## In the words of the source “In April 2021, a security researcher informed Wegmans that a cloud storage container hosted on Microsoft Azure was left unsecured and open to public access, potentially exposing consumers’ sensitive information.” (regulator filing, [New York Attorney General press release: Attorney General James Secures $400,000 from Wegmans After Data Breach Exposed Consumers’ Personal Information](https://ag.ny.gov/press-release/2022/attorney-general-james-secures-400000-wegmans-after-data-breach-exposed-consumers)) ## Primary sources - regulator filing: [New York Attorney General press release: Attorney General James Secures $400,000 from Wegmans After Data Breach Exposed Consumers’ Personal Information](https://ag.ny.gov/press-release/2022/attorney-general-james-secures-400000-wegmans-after-data-breach-exposed-consumers), dated 30 Jun 2022 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Wegmans%20Food%20Markets%2C%20Inc.%20(wegmans-azure-container-2018)) with the entry name in the subject. ## Related pages - [MGM Resorts International 2022: opc](/incidents/mgm-resorts-opc-2019/) - [GeniusU Pte. Ltd. 2022: pdpc](/incidents/geniusu-pdpc-2021/) - [Heroku (Salesforce) and Travis CI OAuth tokens 2022: travi](/incidents/heroku-travisci-oauth-2022/) ## Sources - New York Attorney General press release: Attorney General James Secures $400,000 from Wegmans After Data Breach Exposed Consumers’ Personal Information: https://ag.ny.gov/press-release/2022/attorney-general-james-secures-400000-wegmans-after-data-breach-exposed-consumers (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/wegmans-azure-container-2018/ --- # MGM Resorts International (2022): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. MGM Resorts International disclosed the incident on 19 May 2022. What was exposed, as the sources state it: The OPC finding states MGM confirmed 1,934,090 Canadians were affected, including 5,635 whose government identifier (such as passport, Nexus, health card or military identification number) was compromised; the data set held names, dates of birth, phone numbers, emails and residence addresses. - **Disclosed:** 19 May 2022 - **Incident window:** about 7 July 2019 (access); offered for sale from about 10 July 2019 - **Cloud or provider:** not public - **What was exposed:** The OPC finding states MGM confirmed 1,934,090 Canadians were affected, including 5,635 whose government identifier (such as passport, Nexus, health card or military identification number) was compromised; the data set held names, dates of birth, phone numbers, emails and residence addresses. - **How access was gained:** The OPC finding, relaying MGM's account, says an unauthorized actor logged into a third-party platform used by MGM software developers with an MGM employee's credentials compromised in an earlier unrelated breach, exploited a coding flaw on that platform to bypass multi-factor authentication, found a temporary access token, and used it to reach guest data on an external cloud server for about one hour. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/); also [Third-party access](/access-paths/third-party-access/) - **Kind of control that addresses it:** require phishing-resistant MFA that cannot be bypassed on any platform developers use, block reuse of breached passwords, and limit and expire access tokens for cloud data stores ## In the words of the source “MGM explained that, based on its investigation of the breach, it determined that on approximately 7 July 2019, an unauthorized third party gained access to an external cloud server containing certain MGM guest data.” (regulator filing, [OPC PIPEDA Findings #2022-004: Investigation into MGM breach highlights how to assess risk, and need for timely assessment](https://www.priv.gc.ca/en/opc-actions-and-decisions/investigations/investigations-into-businesses/2022/pipeda-2022-004/)) ## Primary sources - regulator filing: [OPC PIPEDA Findings #2022-004: Investigation into MGM breach highlights how to assess risk, and need for timely assessment](https://www.priv.gc.ca/en/opc-actions-and-decisions/investigations/investigations-into-businesses/2022/pipeda-2022-004/), dated 19 May 2022 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20MGM%20Resorts%20International%20(mgm-resorts-opc-2019)) with the entry name in the subject. ## Related pages - [GeniusU Pte. Ltd. 2022: pdpc](/incidents/geniusu-pdpc-2021/) - [Heroku (Salesforce) and Travis CI OAuth tokens 2022: travi](/incidents/heroku-travisci-oauth-2022/) - [Okta (contractor Sitel) 2022: third-party access](/incidents/okta-sitel-2022/) ## Sources - OPC PIPEDA Findings #2022-004: Investigation into MGM breach highlights how to assess risk, and need for timely assessment: https://www.priv.gc.ca/en/opc-actions-and-decisions/investigations/investigations-into-businesses/2022/pipeda-2022-004/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/mgm-resorts-opc-2019/ --- # GeniusU Pte. Ltd. (2022): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. GeniusU Pte. Ltd. disclosed the incident on 21 Apr 2022. What was exposed, as the sources state it: The PDPC decision states about 1.26 million users' first and last name, email address, location and last sign-in IP address were accessed and exfiltrated from a staging application database. - **Disclosed:** 21 Apr 2022 - **Incident window:** before 12 January 2021 - **Cloud or provider:** GitHub - **What was exposed:** The PDPC decision states about 1.26 million users' first and last name, email address, location and last sign-in IP address were accessed and exfiltrated from a staging application database. - **How access was gained:** The decision says the likely cause was a compromise of a developer's GitHub password, which let the threat actor into the company's GitHub environment; database login credentials had been stored in the codebase there, so the actor reached and exfiltrated the database. The staging database held real personal data. - **Access path:** [Leaked access keys and secrets](/access-paths/leaked-access-key/); also [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** keep database credentials out of code repositories, require MFA on every developer account, and keep real personal data out of test environments ## In the words of the source “As the Organisation had stored the login credentials to the Database in the codebase in its GitHub environment, the threat actor was able to gain access to and exfiltrate personal data stored in the Database.” (regulator filing, [PDPC decision: GeniusU Pte. Ltd. (summary of the decision)](https://www.pdpc.gov.sg/assets/18345b54-0db6-468a-b967-f83664d09ec2)) ## Primary sources - regulator filing: [PDPC decision: GeniusU Pte. Ltd. (summary of the decision)](https://www.pdpc.gov.sg/assets/18345b54-0db6-468a-b967-f83664d09ec2), dated 21 Apr 2022 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20GeniusU%20Pte.%20Ltd.%20(geniusu-pdpc-2021)) with the entry name in the subject. ## Related pages - [Heroku (Salesforce) and Travis CI OAuth tokens 2022: travi](/incidents/heroku-travisci-oauth-2022/) - [Okta (contractor Sitel) 2022: third-party access](/incidents/okta-sitel-2022/) - [Microsoft 2022: dev 0537](/incidents/microsoft-dev-0537-2022/) ## Sources - PDPC decision: GeniusU Pte. Ltd. (summary of the decision): https://www.pdpc.gov.sg/assets/18345b54-0db6-468a-b967-f83664d09ec2 (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/geniusu-pdpc-2021/ --- # Heroku (Salesforce) and Travis CI OAuth tokens (2022): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Heroku (Salesforce) and Travis CI OAuth tokens disclosed the incident on 15 Apr 2022. The ledger reads the access path as third-party access with leaked access keys and secrets. The cloud or provider involved: GitHub. Every field rests on the primary sources listed below. - **Disclosed:** 15 Apr 2022 - **Incident window:** April 2022 (Heroku: 2022-04-07 to 2022-04-13) - **Cloud or provider:** GitHub - **What was exposed:** GitHub stated that stolen OAuth user tokens issued to Heroku and Travis CI were used to download data from dozens of organizations, including npm; Heroku stated the actor also downloaded stored customer GitHub integration tokens and a database of pipeline-level config variables. - **How access was gained:** Heroku stated that initial access to its environment was obtained by using a compromised token for a Heroku machine account, after which stored customer GitHub integration OAuth tokens were downloaded; GitHub stated the tokens were then abused against GitHub.com. - **Access path:** [Third-party access](/access-paths/third-party-access/); also [Leaked access keys and secrets](/access-paths/leaked-access-key/) - **Kind of control that addresses it:** scope and shorten the lifetime of machine-account and integration tokens, store third-party OAuth tokens encrypted with separate keys, and prefer sender-constrained tokens ## In the words of the source “We began investigating how the threat actor gained initial access to the environment and determined it was obtained by leveraging a compromised token for a Heroku machine account.” (company statement, [April 2022 Incident Review](https://blog.heroku.com/april-2022-incident-review)) ## Primary sources - company statement: [April 2022 Incident Review](https://blog.heroku.com/april-2022-incident-review), dated not stated - company statement: [Security alert: Attack campaign involving stolen OAuth user tokens issued to two third-party integrators](https://github.blog/2022-04-15-security-alert-stolen-oauth-user-tokens/), dated 15 Apr 2022 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Heroku%20(Salesforce)%20and%20Travis%20CI%20OAuth%20tokens%20(heroku-travisci-oauth-2022)) with the entry name in the subject. ## Related pages - [Okta (contractor Sitel) 2022: third-party access](/incidents/okta-sitel-2022/) - [Microsoft 2022: dev 0537](/incidents/microsoft-dev-0537-2022/) - [Alfi, Inc. 2022: former executives cloud lockout](/incidents/alfi-former-executives-cloud-lockout-2022/) ## Sources - April 2022 Incident Review: https://blog.heroku.com/april-2022-incident-review (read 16 Nov 2026) - Security alert: Attack campaign involving stolen OAuth user tokens issued to two third-party integrators: https://github.blog/2022-04-15-security-alert-stolen-oauth-user-tokens/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/heroku-travisci-oauth-2022/ --- # Okta (contractor Sitel) (2022): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Okta (contractor Sitel) disclosed the incident on 22 Mar 2022. What was exposed, as the sources state it: The post-mortem says screenshots were taken from a Sitel support engineer's computer; it states a maximum potential impact of 366 customers (approximately 2.5%) and that the Okta service was not breached. - **Disclosed:** 22 Mar 2022 - **Incident window:** January 2022 (alert raised 2022-01-20) - **Cloud or provider:** Okta - **What was exposed:** The post-mortem says screenshots were taken from a Sitel support engineer's computer; it states a maximum potential impact of 366 customers (approximately 2.5%) and that the Okta service was not breached. - **How access was gained:** The post-mortem says an attacker obtained remote access, using RDP, to a computer owned and managed by Okta's sub-processor Sitel and used by a customer support engineer; the attacker did not take over an Okta account. - **Access path:** [Third-party access](/access-paths/third-party-access/) - **Kind of control that addresses it:** require contractor-managed endpoints to meet the same hardening as employer-managed ones, and block remote-desktop access from the internet to devices that hold sessions to the identity provider ## In the words of the source “Our investigation determined that the screenshots, which were not contained in the Sitel summary report, were taken from a Sitel support engineer’s computer upon which an attacker had obtained remote access using RDP.” (provider post mortem, [Okta's Investigation of the January 2022 Compromise](https://www.okta.com/blog/2022/03/oktas-investigation-of-the-january-2022-compromise/)) ## Primary sources - provider post mortem: [Okta's Investigation of the January 2022 Compromise](https://www.okta.com/blog/2022/03/oktas-investigation-of-the-january-2022-compromise/), dated 23 Mar 2022 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Okta%20(contractor%20Sitel)%20(okta-sitel-2022)) with the entry name in the subject. ## Related pages - [Microsoft 2022: dev 0537](/incidents/microsoft-dev-0537-2022/) - [Alfi, Inc. 2022: former executives cloud lockout](/incidents/alfi-former-executives-cloud-lockout-2022/) - [A publicly traded wireless communications manufacturer (ca](/incidents/ubiquiti-aws-github-2020/) ## Sources - Okta's Investigation of the January 2022 Compromise: https://www.okta.com/blog/2022/03/oktas-investigation-of-the-january-2022-compromise/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/okta-sitel-2022/ --- # Microsoft (2022): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Microsoft disclosed the incident on 22 Mar 2022. What was exposed, as the sources state it: The company stated that the actor claimed to have exfiltrated portions of source code, that no customer code or data was involved, and that a single account had been compromised, granting limited access. - **Disclosed:** 22 Mar 2022 - **Incident window:** not public - **Cloud or provider:** not public - **What was exposed:** The company stated that the actor claimed to have exfiltrated portions of source code, that no customer code or data was involved, and that a single account had been compromised, granting limited access. - **How access was gained:** The company stated that a single account was compromised, granting limited access, and that the tactics used in this intrusion reflect those described in the same post (including stolen credentials and MFA prompt abuse). - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** require phishing-resistant MFA, block MFA approval prompts and session-token replay, and limit what a single account can reach ## In the words of the source “Our investigation has found a single account had been compromised, granting limited access.” (provider post mortem, [DEV-0537 criminal actor targeting organizations for data exfiltration and destruction](https://www.microsoft.com/en-us/security/blog/2022/03/22/dev-0537-criminal-actor-targeting-organizations-for-data-exfiltration-and-destruction/)) ## Primary sources - provider post mortem: [DEV-0537 criminal actor targeting organizations for data exfiltration and destruction](https://www.microsoft.com/en-us/security/blog/2022/03/22/dev-0537-criminal-actor-targeting-organizations-for-data-exfiltration-and-destruction/), dated 22 Mar 2022 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Microsoft%20(microsoft-dev-0537-2022)) with the entry name in the subject. ## Related pages - [Alfi, Inc. 2022: former executives cloud lockout](/incidents/alfi-former-executives-cloud-lockout-2022/) - [A publicly traded wireless communications manufacturer (ca](/incidents/ubiquiti-aws-github-2020/) - [GoDaddy 2021: managed wordpress](/incidents/godaddy-managed-wordpress-2021/) ## Sources - DEV-0537 criminal actor targeting organizations for data exfiltration and destruction: https://www.microsoft.com/en-us/security/blog/2022/03/22/dev-0537-criminal-actor-targeting-organizations-for-data-exfiltration-and-destruction/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/microsoft-dev-0537-2022/ --- # Alfi, Inc. (2022): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Alfi, Inc. disclosed the incident on 23 Feb 2022. The ledger reads the access path as retained access after departure or change. The cloud or provider involved: Amazon Web Services. Every field rests on the primary sources listed below. Fields no primary source states read not public. - **Disclosed:** 23 Feb 2022 - **Incident window:** After the executives were placed on leave on October 22, 2021 (the dates of the access are not stated) - **Cloud or provider:** Amazon Web Services - **What was exposed:** The filing says the company's Google Workspace (email accounts, calendars, contacts, chats and saved files) and its Amazon Web Services account (which hosts tablet data and a software development platform) were taken over so the company could not administer them, that downloadable access to all data in the Google Workspace was obtained from Google, and that files were accessed and downloaded from the company's Google Drive. The filing says the company continued business through workarounds. - **How access was gained:** The filing says that after the executives were placed on leave and told not to access the company's computer systems or platforms, one of them took control of the Google Workspace and Amazon Web Services accounts, and two others accessed and downloaded files from Google Drive. - **Access path:** [Retained access after departure or change](/access-paths/retained-access/) - **Kind of control that addresses it:** revoke administrator and root credentials, cloud console access and data export rights on the day an executive or administrator is placed on leave or leaves, and keep break-glass admin accounts under separate control ## In the words of the source “After being placed on leave and having been instructed at such time by the Company not to access its computer systems or platforms:” (regulator filing, [Alfi, Inc. Form 8-K, Item 8.01, filed 2022-02-23](https://www.sec.gov/Archives/edgar/data/1833908/000110465922026150/tm227500d1_8k.htm)) ## Primary sources - regulator filing: [Alfi, Inc. Form 8-K, Item 8.01, filed 2022-02-23](https://www.sec.gov/Archives/edgar/data/1833908/000110465922026150/tm227500d1_8k.htm), dated 23 Feb 2022 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Alfi%2C%20Inc.%20(alfi-former-executives-cloud-lockout-2022)) with the entry name in the subject. ## Related pages - [A publicly traded wireless communications manufacturer (ca](/incidents/ubiquiti-aws-github-2020/) - [GoDaddy 2021: managed wordpress](/incidents/godaddy-managed-wordpress-2021/) - [Commeasure Pte Ltd (operates the RedDoorz booking platform](/incidents/commeasure-reddoorz-pdpc-2020/) ## Sources - Alfi, Inc. Form 8-K, Item 8.01, filed 2022-02-23: https://www.sec.gov/Archives/edgar/data/1833908/000110465922026150/tm227500d1_8k.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/alfi-former-executives-cloud-lockout-2022/ --- # A publicly traded wireless communications manufacturer (called Company-1 in the U.S. Attorney release) (2021): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. A publicly traded wireless communications manufacturer (called Company-1 in the U.S. Attorney release) disclosed the incident on 1 Dec 2021. What was exposed, as the sources state it: The release says gigabytes of confidential data were downloaded from the company’s AWS and GitHub infrastructure and later used in an extortion attempt for millions of dollars. - **Disclosed:** 1 Dec 2021 - **Incident window:** about December 2020, as the release says - **Cloud or provider:** AWS - **What was exposed:** The release says gigabytes of confidential data were downloaded from the company’s AWS and GitHub infrastructure and later used in an extortion attempt for millions of dollars. - **How access was gained:** The release says an employee who was a senior developer held credentials for the company’s AWS and GitHub servers and repeatedly misused his administrative access to download the data; the company’s own account of an outside intruder was, the release says, false. - **Access path:** [Over-permissioned identities](/access-paths/over-permissioned-identity/) - **Kind of control that addresses it:** limit cloud administrator and source-control privileges to the smallest set of people, require approval for bulk exports, and alert on large downloads by administrative identities ## In the words of the source “a senior developer who had access to credentials for Company-1’s Amazon Web Services (AWS) and GitHub Inc. (GitHub) servers.” (court document, [U.S. Attorney SDNY: Former Employee Of Technology Company Charged With Stealing Confidential Data And Extorting Company For Ransom](https://web.archive.org/web/2024/https://www.justice.gov/usao-sdny/pr/former-employee-technology-company-charged-stealing-confidential-data-and-extorting)) ## Primary sources - court document: [U.S. Attorney SDNY: Former Employee Of Technology Company Charged With Stealing Confidential Data And Extorting Company For Ransom](https://web.archive.org/web/2024/https://www.justice.gov/usao-sdny/pr/former-employee-technology-company-charged-stealing-confidential-data-and-extorting), dated 1 Dec 2021 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20A%20publicly%20traded%20wireless%20communications%20manufacturer%20(called%20Company-1%20in%20the%20U.S.%20Attorney%20release)%20(ubiquiti-aws-github-2020)) with the entry name in the subject. ## Related pages - [GoDaddy 2021: managed wordpress](/incidents/godaddy-managed-wordpress-2021/) - [Commeasure Pte Ltd (operates the RedDoorz booking platform](/incidents/commeasure-reddoorz-pdpc-2020/) - [Stylez Pte Ltd 2021: icompare pdpc](/incidents/stylez-icompare-pdpc-2019/) ## Sources - U.S. Attorney SDNY: Former Employee Of Technology Company Charged With Stealing Confidential Data And Extorting Company For Ransom: https://web.archive.org/web/2024/https://www.justice.gov/usao-sdny/pr/former-employee-technology-company-charged-stealing-confidential-data-and-extorting (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/ubiquiti-aws-github-2020/ --- # GoDaddy (2021): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. GoDaddy disclosed the incident on 22 Nov 2021. What was exposed, as the sources state it: The company stated that up to 1.2 million active and inactive Managed WordPress customers had email address and customer number exposed, along with the original WordPress admin password, sFTP and database credentials for active customers, and the SSL private key for a subset. - **Disclosed:** 22 Nov 2021 - **Incident window:** 2021-09-06 to 2021-11-17 - **Cloud or provider:** GoDaddy Managed WordPress hosting - **What was exposed:** The company stated that up to 1.2 million active and inactive Managed WordPress customers had email address and customer number exposed, along with the original WordPress admin password, sFTP and database credentials for active customers, and the SSL private key for a subset. - **How access was gained:** The company stated that an unauthorized third party used a compromised password to access the provisioning system in the legacy code base for Managed WordPress. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** require MFA on provisioning and administrative systems, including legacy ones, and avoid retaining reusable customer credentials after provisioning ## In the words of the source “Using a compromised password, an unauthorized third party accessed the provisioning system in our legacy code base for Managed WordPress.” (company statement, [GoDaddy Announces Security Incident Affecting Managed WordPress Service (8-K Exhibit 99.1)](https://www.sec.gov/Archives/edgar/data/1609711/000160971121000122/gddyblogpostnov222021.htm)) ## Primary sources - company statement: [GoDaddy Announces Security Incident Affecting Managed WordPress Service (8-K Exhibit 99.1)](https://www.sec.gov/Archives/edgar/data/1609711/000160971121000122/gddyblogpostnov222021.htm), dated 22 Nov 2021 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20GoDaddy%20(godaddy-managed-wordpress-2021)) with the entry name in the subject. ## Related pages - [Commeasure Pte Ltd (operates the RedDoorz booking platform](/incidents/commeasure-reddoorz-pdpc-2020/) - [Stylez Pte Ltd 2021: icompare pdpc](/incidents/stylez-icompare-pdpc-2019/) - [The National Kidney Foundation 2021: pdpc](/incidents/national-kidney-foundation-pdpc-2020/) ## Sources - GoDaddy Announces Security Incident Affecting Managed WordPress Service (8-K Exhibit 99.1): https://www.sec.gov/Archives/edgar/data/1609711/000160971121000122/gddyblogpostnov222021.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/godaddy-managed-wordpress-2021/ --- # Commeasure Pte Ltd (operates the RedDoorz booking platform) (2021): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Commeasure Pte Ltd (operates the RedDoorz booking platform) disclosed the incident on 11 Nov 2021. What was exposed, as the sources state it: The PDPC decision states a database of 5,892,843 customer records (name, contact number, email, date of birth, hashed password and booking information) was accessed and exfiltrated; masked credit card numbers were not accessed. - **Disclosed:** 11 Nov 2021 - **Incident window:** September 2020 (customers notified 26 September 2020) - **Cloud or provider:** AWS - **What was exposed:** The PDPC decision states a database of 5,892,843 customer records (name, contact number, email, date of birth, hashed password and booking information) was accessed and exfiltrated; masked credit card numbers were not accessed. - **How access was gained:** The decision says the threat actors most likely obtained an AWS access key embedded in an Android application package, created in 2015, last updated in January 2018, still downloadable from the Google Play Store, and marked as a test key although it could reach a live production database in Amazon RDS. - **Access path:** [Leaked access keys and secrets](/access-paths/leaked-access-key/) - **Kind of control that addresses it:** keep cloud access keys out of mobile app packages and code, scope keys to the least access needed, and retire old app builds and the keys inside them ## In the words of the source “The AWS 3 access key was embedded within an Android application package ( the affected APK ) publicly available for download from the Google Play Store.” (regulator filing, [PDPC decision: Commeasure Pte Ltd (15 September 2021)](https://www.pdpc.gov.sg/assets/f78cffe6-5efe-4757-b141-2d273702b812)) ## Primary sources - regulator filing: [PDPC decision: Commeasure Pte Ltd (15 September 2021)](https://www.pdpc.gov.sg/assets/f78cffe6-5efe-4757-b141-2d273702b812), dated 11 Nov 2021 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Commeasure%20Pte%20Ltd%20(operates%20the%20RedDoorz%20booking%20platform)%20(commeasure-reddoorz-pdpc-2020)) with the entry name in the subject. ## Related pages - [Stylez Pte Ltd 2021: icompare pdpc](/incidents/stylez-icompare-pdpc-2019/) - [The National Kidney Foundation 2021: pdpc](/incidents/national-kidney-foundation-pdpc-2020/) - [Sendtech Pte. Ltd. 2021: pdpc](/incidents/sendtech-pdpc-2021/) ## Sources - PDPC decision: Commeasure Pte Ltd (15 September 2021): https://www.pdpc.gov.sg/assets/f78cffe6-5efe-4757-b141-2d273702b812 (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/commeasure-reddoorz-pdpc-2020/ --- # Stylez Pte Ltd (2021): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Stylez Pte Ltd disclosed the incident on 14 Oct 2021. What was exposed, as the sources state it: The PDPC decision states a testing database (a text file of renovation and interior design clients, 2009 to 2016) exposing the name, email address and phone number of 9,983 individuals was uploaded to the dark web. - **Disclosed:** 14 Oct 2021 - **Incident window:** before December 2019 - **Cloud or provider:** not public - **What was exposed:** The PDPC decision states a testing database (a text file of renovation and interior design clients, 2009 to 2016) exposing the name, email address and phone number of 9,983 individuals was uploaded to the dark web. - **How access was gained:** The decision says the testing database was hosted on a cloud server leased from a cloud storage service provider, stored in a publicly accessible directory without access controls, so it was reachable from the internet and was crawled and indexed by search engines. - **Access path:** [Public storage, databases and services](/access-paths/public-resource/) - **Kind of control that addresses it:** do not keep copies of production data in test environments, put access controls on every web-reachable directory, and scan for exposed files ## In the words of the source “Firstly, the Testing Database was stored in a publicly accessible directory in the Server without any access controls.” (regulator filing, [PDPC decision: Stylez Pte Ltd (iCompare.sg)](https://www.pdpc.gov.sg/assets/37845a6b-efc4-426d-bc1f-58e4e7da59ff)) ## Primary sources - regulator filing: [PDPC decision: Stylez Pte Ltd (iCompare.sg)](https://www.pdpc.gov.sg/assets/37845a6b-efc4-426d-bc1f-58e4e7da59ff), dated 14 Oct 2021 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Stylez%20Pte%20Ltd%20(stylez-icompare-pdpc-2019)) with the entry name in the subject. ## Related pages - [The National Kidney Foundation 2021: pdpc](/incidents/national-kidney-foundation-pdpc-2020/) - [Sendtech Pte. Ltd. 2021: pdpc](/incidents/sendtech-pdpc-2021/) - [Cetera Entities (Cetera Advisor Networks LLC, Cetera Inves](/incidents/cetera-email-account-takeovers-2017/) ## Sources - PDPC decision: Stylez Pte Ltd (iCompare.sg): https://www.pdpc.gov.sg/assets/37845a6b-efc4-426d-bc1f-58e4e7da59ff (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/stylez-icompare-pdpc-2019/ --- # The National Kidney Foundation (2021): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. The National Kidney Foundation disclosed the incident on 14 Oct 2021. The ledger reads the access path as stolen user credentials and session tokens. The cloud or provider involved: Microsoft 365. Every field rests on the primary sources listed below. - **Disclosed:** 14 Oct 2021 - **Incident window:** 14 to 17 May 2020 - **Cloud or provider:** Microsoft 365 - **What was exposed:** The PDPC decision states the mailbox of one employee held 23,145 emails with the personal data of approximately 500 individuals (patients, employees and third parties), including bank account numbers, dialysis readings and medical conditions. It states the mailbox was believed to have been downloaded in full. - **How access was gained:** The decision says an employee received a phishing email linking to a website that asked for account credentials; the hacker is believed to have obtained them, accessed the employee's mailbox, which was hosted on Microsoft Office 365 and reachable by webmail, synchronised it on 17 May 2020, and used it to send phishing emails to 1,039 external contacts. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** require phishing-resistant MFA for web mail, restrict mailbox access to managed devices, and keep sensitive records out of mailboxes ## In the words of the source “Investigations revealed that, on 14 May 2020, Employee A received a phishing email containing a hyperlink to a website with a further link to another website seeking his account credentials.” (regulator filing, [PDPC decision: The National Kidney Foundation](https://www.pdpc.gov.sg/assets/8238cce1-16fd-4cec-b9ab-ffd8e8445a7b)) ## Primary sources - regulator filing: [PDPC decision: The National Kidney Foundation](https://www.pdpc.gov.sg/assets/8238cce1-16fd-4cec-b9ab-ffd8e8445a7b), dated 14 Oct 2021 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20The%20National%20Kidney%20Foundation%20(national-kidney-foundation-pdpc-2020)) with the entry name in the subject. ## Related pages - [Sendtech Pte. Ltd. 2021: pdpc](/incidents/sendtech-pdpc-2021/) - [Cetera Entities (Cetera Advisor Networks LLC, Cetera Inves](/incidents/cetera-email-account-takeovers-2017/) - [Cambridge Investment Research Inc. and Cambridge Investmen](/incidents/cambridge-email-account-takeovers-2018/) ## Sources - PDPC decision: The National Kidney Foundation: https://www.pdpc.gov.sg/assets/8238cce1-16fd-4cec-b9ab-ffd8e8445a7b (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/national-kidney-foundation-pdpc-2020/ --- # Sendtech Pte. Ltd. (2021): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Sendtech Pte. Ltd. disclosed the incident on 21 Sep 2021. The ledger reads the access path as leaked access keys and secrets with retained access after departure or change. The cloud or provider involved: AWS. Every field rests on the primary sources listed below. - **Disclosed:** 21 Sep 2021 - **Incident window:** access key created in 2015 and not rotated since; discovered 10 February 2021 - **Cloud or provider:** AWS - **What was exposed:** The PDPC decision states the personal data of 64,196 customers and 3,401 contractors and the contractors' employees was accessed: for customers the email address, contact number, home address and last four digits of the card; for contractors and their employees a profile photo and copies of the NRIC or work permit. It states there was no evidence of data exfiltration. - **How access was gained:** The decision says the cause was a compromised AWS access key created in 2015 and not rotated since. With it the attacker gained admin privileges, created another admin account and queried the buckets storing personal data. The company suspected the key could have been compromised through former or current employees but the decision does not confirm how. - **Access path:** [Leaked access keys and secrets](/access-paths/leaked-access-key/); also [Retained access after departure or change](/access-paths/retained-access/) - **Kind of control that addresses it:** rotate long-lived cloud access keys on a schedule, remove keys that former developers could have held, and alert on new administrator accounts created with a key ## In the words of the source “The cause of Incident was a compromised AWS access key.” (regulator filing, [PDPC decision: Sendtech Pte. Ltd. (DP-2102-B7884)](https://www.pdpc.gov.sg/assets/51342b7d-edaf-484b-82d5-e0dec632f4fc)) ## Primary sources - regulator filing: [PDPC decision: Sendtech Pte. Ltd. (DP-2102-B7884)](https://www.pdpc.gov.sg/assets/51342b7d-edaf-484b-82d5-e0dec632f4fc), dated 21 Sep 2021 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Sendtech%20Pte.%20Ltd.%20(sendtech-pdpc-2021)) with the entry name in the subject. ## Related pages - [Cetera Entities (Cetera Advisor Networks LLC, Cetera Inves](/incidents/cetera-email-account-takeovers-2017/) - [Cambridge Investment Research Inc. and Cambridge Investmen](/incidents/cambridge-email-account-takeovers-2018/) - [KMS Financial Services Inc. 2021: email account takeovers](/incidents/kms-financial-email-account-takeovers-2018/) ## Sources - PDPC decision: Sendtech Pte. Ltd. (DP-2102-B7884): https://www.pdpc.gov.sg/assets/51342b7d-edaf-484b-82d5-e0dec632f4fc (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/sendtech-pdpc-2021/ --- # Cetera Entities (Cetera Advisor Networks LLC, Cetera Investment Services LLC, Cetera Financial Specialists LLC, Cetera Advisors LLC, Cetera Investment Advisers LLC) (2021): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Cetera Entities (Cetera Advisor Networks LLC, Cetera Investment Services LLC, Cetera Financial Specialists LLC, Cetera Advisors LLC, Cetera Investment Advisers LLC) disclosed the incident on 30 Aug 2021. What was exposed, as the sources state it: The SEC states personally identifying information of at least 4,388 customers and clients was exposed. - **Disclosed:** 30 Aug 2021 - **Incident window:** November 2017 to June 2020 - **Cloud or provider:** not public - **What was exposed:** The SEC states personally identifying information of at least 4,388 customers and clients was exposed. - **How access was gained:** The SEC states the cloud-based email accounts of over 60 personnel were taken over by unauthorized third parties, and that none of the taken over accounts were protected in a manner consistent with the firms’ policies. The press release does not name the email provider or state how the accounts were taken over. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** require phishing-resistant multifactor authentication on every cloud-hosted email account and block legacy sign-in methods that bypass it ## In the words of the source “between November 2017 and June 2020, cloud-based email accounts of over 60 Cetera Entities' personnel were taken over by unauthorized third parties” (regulator filing, [SEC press release 2021-169: SEC Announces Three Actions Charging Deficient Cybersecurity Procedures](https://www.sec.gov/newsroom/press-releases/2021-169)) ## Primary sources - regulator filing: [SEC press release 2021-169: SEC Announces Three Actions Charging Deficient Cybersecurity Procedures](https://www.sec.gov/newsroom/press-releases/2021-169), dated 30 Aug 2021 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Cetera%20Entities%20(Cetera%20Advisor%20Networks%20LLC%2C%20Cetera%20Investment%20Services%20LLC%2C%20Cetera%20Financial%20Specialists%20LLC%2C%20Cetera%20Advisors%20LLC%2C%20Cetera%20Investment%20Advisers%20LLC)%20(cetera-email-account-takeovers-2017)) with the entry name in the subject. ## Related pages - [Cambridge Investment Research Inc. and Cambridge Investmen](/incidents/cambridge-email-account-takeovers-2018/) - [KMS Financial Services Inc. 2021: email account takeovers](/incidents/kms-financial-email-account-takeovers-2018/) - [Microsoft (Azure Cosmos DB) 2021: cosmosdb chaosdb](/incidents/azure-cosmosdb-chaosdb-2021/) ## Sources - SEC press release 2021-169: SEC Announces Three Actions Charging Deficient Cybersecurity Procedures: https://www.sec.gov/newsroom/press-releases/2021-169 (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/cetera-email-account-takeovers-2017/ --- # Cambridge Investment Research Inc. and Cambridge Investment Research Advisors Inc. (2021): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Cambridge Investment Research Inc. and Cambridge Investment Research Advisors Inc. disclosed the incident on 30 Aug 2021. What was exposed, as the sources state it: The SEC states personally identifying information of at least 2,177 customers and clients was exposed. - **Disclosed:** 30 Aug 2021 - **Incident window:** January 2018 to July 2021 - **Cloud or provider:** not public - **What was exposed:** The SEC states personally identifying information of at least 2,177 customers and clients was exposed. - **How access was gained:** The SEC states the cloud-based email accounts of over 121 representatives were taken over by unauthorized third parties. The SEC order finds the firm discovered the first takeover in January 2018 but did not adopt and implement firm-wide enhanced security measures for those accounts until 2021. The press release does not name the email provider or state how the accounts were taken over. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** require phishing-resistant multifactor authentication on every cloud-hosted email account and block legacy sign-in methods that bypass it ## In the words of the source “between January 2018 and July 2021, cloud-based email accounts of over 121 Cambridge representatives were taken over by unauthorized third parties” (regulator filing, [SEC press release 2021-169: SEC Announces Three Actions Charging Deficient Cybersecurity Procedures](https://www.sec.gov/newsroom/press-releases/2021-169)) ## Primary sources - regulator filing: [SEC press release 2021-169: SEC Announces Three Actions Charging Deficient Cybersecurity Procedures](https://www.sec.gov/newsroom/press-releases/2021-169), dated 30 Aug 2021 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Cambridge%20Investment%20Research%20Inc.%20and%20Cambridge%20Investment%20Research%20Advisors%20Inc.%20(cambridge-email-account-takeovers-2018)) with the entry name in the subject. ## Related pages - [KMS Financial Services Inc. 2021: email account takeovers](/incidents/kms-financial-email-account-takeovers-2018/) - [Microsoft (Azure Cosmos DB) 2021: cosmosdb chaosdb](/incidents/azure-cosmosdb-chaosdb-2021/) - [Codecov 2021: leaked access keys and secrets](/incidents/codecov-2021/) ## Sources - SEC press release 2021-169: SEC Announces Three Actions Charging Deficient Cybersecurity Procedures: https://www.sec.gov/newsroom/press-releases/2021-169 (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/cambridge-email-account-takeovers-2018/ --- # KMS Financial Services Inc. (2021): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. KMS Financial Services Inc. disclosed the incident on 30 Aug 2021. What was exposed, as the sources state it: The SEC states personally identifying information of approximately 4,900 customers and clients was exposed. The ledger reads the access path as stolen user credentials and session tokens. - **Disclosed:** 30 Aug 2021 - **Incident window:** September 2018 to December 2019 - **Cloud or provider:** not public - **What was exposed:** The SEC states personally identifying information of approximately 4,900 customers and clients was exposed. - **How access was gained:** The SEC states the cloud-based email accounts of 15 financial advisers or their assistants were taken over by unauthorized third parties. The SEC order finds the firm did not adopt written policies requiring additional firm-wide security measures until May 2020 and did not fully implement them firm-wide until August 2020. The press release does not name the email provider or state how the accounts were taken over. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** require phishing-resistant multifactor authentication on every cloud-hosted email account and block legacy sign-in methods that bypass it ## In the words of the source “between September 2018 and December 2019, cloud-based email accounts of 15 KMS financial advisers or their assistants were taken over by unauthorized third parties” (regulator filing, [SEC press release 2021-169: SEC Announces Three Actions Charging Deficient Cybersecurity Procedures](https://www.sec.gov/newsroom/press-releases/2021-169)) ## Primary sources - regulator filing: [SEC press release 2021-169: SEC Announces Three Actions Charging Deficient Cybersecurity Procedures](https://www.sec.gov/newsroom/press-releases/2021-169), dated 30 Aug 2021 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20KMS%20Financial%20Services%20Inc.%20(kms-financial-email-account-takeovers-2018)) with the entry name in the subject. ## Related pages - [Microsoft (Azure Cosmos DB) 2021: cosmosdb chaosdb](/incidents/azure-cosmosdb-chaosdb-2021/) - [Codecov 2021: leaked access keys and secrets](/incidents/codecov-2021/) - [Mimecast 2021: solarwinds certificate](/incidents/mimecast-solarwinds-certificate-2021/) ## Sources - SEC press release 2021-169: SEC Announces Three Actions Charging Deficient Cybersecurity Procedures: https://www.sec.gov/newsroom/press-releases/2021-169 (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/kms-financial-email-account-takeovers-2018/ --- # Microsoft (Azure Cosmos DB) (2021): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Microsoft (Azure Cosmos DB) disclosed the incident on 27 Aug 2021. What was exposed, as the sources state it: The post-mortem says primary read-write keys of a subset of Azure Cosmos DB customers could be reached, customers were notified to regenerate keys, and no customer data was accessed by third parties or security researchers. - **Disclosed:** 27 Aug 2021 - **Incident window:** Researcher activity between 2021-08-07 and 2021-08-13 - **Cloud or provider:** Microsoft Azure - **What was exposed:** The post-mortem says primary read-write keys of a subset of Azure Cosmos DB customers could be reached, customers were notified to regenerate keys, and no customer data was accessed by third parties or security researchers. - **How access was gained:** The post-mortem says a vulnerability in the Azure Cosmos DB Jupyter Notebook feature could allow a user to gain access to another customer's resources by using the account's primary read-write key. - **Access path:** [Cross-tenant access](/access-paths/cross-tenant/) - **Kind of control that addresses it:** use role-based access control and network restrictions instead of account-wide keys, and rotate account keys on a schedule ## In the words of the source “a vulnerability in the Azure Cosmos DB Jupyter Notebook feature that could potentially allow a user to gain access to another customer’s resources by using the account’s primary read-write key.” (provider post mortem, [Update on the vulnerability in the Azure Cosmos DB Jupyter Notebook Feature (archived copy)](https://web.archive.org/web/2021/https://msrc-blog.microsoft.com/2021/08/27/update-on-vulnerability-in-the-azure-cosmos-db-jupyter-notebook-feature/)) ## Primary sources - provider post mortem: [Update on the vulnerability in the Azure Cosmos DB Jupyter Notebook Feature (archived copy)](https://web.archive.org/web/2021/https://msrc-blog.microsoft.com/2021/08/27/update-on-vulnerability-in-the-azure-cosmos-db-jupyter-notebook-feature/), dated 27 Aug 2021 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Microsoft%20(Azure%20Cosmos%20DB)%20(azure-cosmosdb-chaosdb-2021)) with the entry name in the subject. ## Related pages - [Codecov 2021: leaked access keys and secrets](/incidents/codecov-2021/) - [Mimecast 2021: solarwinds certificate](/incidents/mimecast-solarwinds-certificate-2021/) - [SkyMed International, Inc. 2020: ftc](/incidents/skymed-ftc-2020/) ## Sources - Update on the vulnerability in the Azure Cosmos DB Jupyter Notebook Feature (archived copy): https://web.archive.org/web/2021/https://msrc-blog.microsoft.com/2021/08/27/update-on-vulnerability-in-the-azure-cosmos-db-jupyter-notebook-feature/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/azure-cosmosdb-chaosdb-2021/ --- # Codecov (2021): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Codecov disclosed the incident on 15 Apr 2021. The ledger reads the access path as leaked access keys and secrets with third-party access. The cloud or provider involved: Codecov. Every field rests on the primary sources listed below. Fields no primary source states read not public. - **Disclosed:** 15 Apr 2021 - **Incident window:** 31 January 2021 to 1 April 2021 - **Cloud or provider:** Codecov - **What was exposed:** The company stated that credentials, tokens or keys that customers passed through their CI runner, and the git remote information of repositories using the Bash Uploaders, could have been exported; it counts 108 windows of time in which the altered script was live between 31 January and 1 April. - **How access was gained:** The company stated that an error in its Docker image creation process allowed the actor to extract the credential required to modify the Bash Uploader script, which was then altered to send CI environment variables to a third-party server. - **Access path:** [Leaked access keys and secrets](/access-paths/leaked-access-key/); also [Third-party access](/access-paths/third-party-access/) - **Kind of control that addresses it:** keep signing and publishing credentials out of build artefacts and container images, scope them to a single task, and verify checksums of scripts fetched into CI ## In the words of the source “The actor gained access because of an error in Codecov’s Docker image creation process that allowed the actor to extract the credential required to modify our Bash Uploader script.” (company statement, [Bash Uploader Security Update](https://about.codecov.io/security-update/)) ## Primary sources - company statement: [Bash Uploader Security Update](https://about.codecov.io/security-update/), dated 15 Apr 2021 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Codecov%20(codecov-2021)) with the entry name in the subject. ## Related pages - [Mimecast 2021: solarwinds certificate](/incidents/mimecast-solarwinds-certificate-2021/) - [SkyMed International, Inc. 2020: ftc](/incidents/skymed-ftc-2020/) - [Ascension Data & Analytics 2020: public storage, databases](/incidents/ascension-data-2019/) ## Sources - Bash Uploader Security Update: https://about.codecov.io/security-update/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/codecov-2021/ --- # Mimecast (2021): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Mimecast disclosed the incident on 16 Mar 2021. The ledger reads the access path as third-party access with leaked access keys and secrets. The cloud or provider involved: Microsoft 365. Every field rests on the primary sources listed below. Fields no primary source states read not public. - **Disclosed:** 16 Mar 2021 - **Incident window:** Became aware in January 2021; forensic investigation completed in March 2021 (per the company) - **Cloud or provider:** Microsoft 365 - **What was exposed:** The company stated the threat actor accessed certain Mimecast-issued certificates and related customer server connection information, a subset of email addresses and other contact information, hashed and salted credentials, and a limited number of source code repositories. It stated Microsoft informed it that a certificate used to authenticate some Mimecast products to Microsoft 365 Exchange Web Services was used to connect to a low single-digit number of mutual customers' Microsoft 365 tenants. It stated it had no evidence that email or archive content held for customers was accessed. - **How access was gained:** The company stated the threat actor used the SolarWinds supply-chain compromise to gain access to part of its production grid environment, and from there accessed Mimecast-issued certificates and related connection information; the certificate was then used against a small number of customer Microsoft 365 tenants. - **Access path:** [Third-party access](/access-paths/third-party-access/); also [Leaked access keys and secrets](/access-paths/leaked-access-key/) - **Kind of control that addresses it:** restrict and monitor vendor-issued certificates and service credentials used to reach customer tenants, rotate them on any supplier compromise, and move to scoped token-based authorization instead of long-lived certificates ## In the words of the source “During our investigation, we learned that the threat actor used the SolarWinds supply-chain compromise to gain access to part of our production grid environment.” (company statement, [Incident Report (Mimecast), published March 16, 2021, archived copy of 2021-03-17](https://web.archive.org/web/20210317013526/https://www.mimecast.com/incident-report/)) ## Primary sources - company statement: [Incident Report (Mimecast), published March 16, 2021, archived copy of 2021-03-17](https://web.archive.org/web/20210317013526/https://www.mimecast.com/incident-report/), dated 16 Mar 2021 - regulator filing: [Mimecast Form 8-K, Exhibit 99.2, filed 2021-03-16](https://www.sec.gov/Archives/edgar/data/1644675/000119312521082200/d141664dex992.htm), dated 16 Mar 2021 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Mimecast%20(mimecast-solarwinds-certificate-2021)) with the entry name in the subject. ## Related pages - [SkyMed International, Inc. 2020: ftc](/incidents/skymed-ftc-2020/) - [Ascension Data & Analytics 2020: public storage, databases](/incidents/ascension-data-2019/) - [SolarWinds 2020: third-party access](/incidents/solarwinds-2020/) ## Sources - Incident Report (Mimecast), published March 16, 2021, archived copy of 2021-03-17: https://web.archive.org/web/20210317013526/https://www.mimecast.com/incident-report/ (read 16 Nov 2026) - Mimecast Form 8-K, Exhibit 99.2, filed 2021-03-16: https://www.sec.gov/Archives/edgar/data/1644675/000119312521082200/d141664dex992.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/mimecast-solarwinds-certificate-2021/ --- # SkyMed International, Inc. (2020): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. SkyMed International, Inc. disclosed the incident on 16 Dec 2020. What was exposed, as the sources state it: The FTC states the company left unsecured a cloud database containing 130,000 membership records. According to the complaint, the database held names, dates of birth, home addresses, health information and membership account numbers in plain text. - **Disclosed:** 16 Dec 2020 - **Incident window:** not public - **Cloud or provider:** not public - **What was exposed:** The FTC states the company left unsecured a cloud database containing 130,000 membership records. According to the complaint, the database held names, dates of birth, home addresses, health information and membership account numbers in plain text. - **How access was gained:** The FTC press release says the database was unsecured and exposed by a security researcher, and that it could be located and accessed by anyone on the Internet. The press release says the company deleted the database after confirming that the data was online and publicly accessible. - **Access path:** [Public storage, databases and services](/access-paths/public-resource/) - **Kind of control that addresses it:** block public access to cloud databases by default, alert on any data store that becomes reachable from the internet, and encrypt personal data at rest ## In the words of the source “The unsecured database, exposed by a security researcher, could be located and accessed by anyone on the Internet” (regulator filing, [FTC press release: Company that Provides Travel Emergency Services Settles FTC Allegations it Failed to Secure Sensitive Consumer Data (web.archive.org copy)](https://web.archive.org/web/2024/https://www.ftc.gov/news-events/news/press-releases/2020/12/company-provides-travel-emergency-services-settles-ftc-allegations-it-failed-secure-sensitive)) ## Primary sources - regulator filing: [FTC press release: Company that Provides Travel Emergency Services Settles FTC Allegations it Failed to Secure Sensitive Consumer Data (web.archive.org copy)](https://web.archive.org/web/2024/https://www.ftc.gov/news-events/news/press-releases/2020/12/company-provides-travel-emergency-services-settles-ftc-allegations-it-failed-secure-sensitive), dated 16 Dec 2020 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20SkyMed%20International%2C%20Inc.%20(skymed-ftc-2020)) with the entry name in the subject. ## Related pages - [Ascension Data & Analytics 2020: public storage, databases](/incidents/ascension-data-2019/) - [SolarWinds 2020: third-party access](/incidents/solarwinds-2020/) - [Chan Brothers Travel Pte Ltd 2020: pdpc](/incidents/chan-brothers-travel-pdpc-2019/) ## Sources - FTC press release: Company that Provides Travel Emergency Services Settles FTC Allegations it Failed to Secure Sensitive Consumer Data (web.archive.org copy): https://web.archive.org/web/2024/https://www.ftc.gov/news-events/news/press-releases/2020/12/company-provides-travel-emergency-services-settles-ftc-allegations-it-failed-secure-sensitive (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/skymed-ftc-2020/ --- # Ascension Data & Analytics (2020): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Ascension Data & Analytics disclosed the incident on 15 Dec 2020. The ledger reads the access path as public storage, databases and services with third-party access. The cloud or provider involved: not public. Every field rests on the primary sources listed below. - **Disclosed:** 15 Dec 2020 - **Incident window:** January 2018 to about January 2019 (per the FTC complaint) - **Cloud or provider:** not public - **What was exposed:** Mortgage documents relating to approximately 37,000 mortgages, containing information about 60,593 consumers: names, dates of birth, Social Security numbers, loan information, credit and debit account numbers, drivers license numbers, credit files and other personal and financial information. The complaint says approximately 52 unauthorized IP addresses accessed the server and storage location. - **How access was gained:** The FTC complaint says a vendor hired for text recognition scanning stored the documents on a cloud-based server and a separate cloud-based storage location, misconfigured both, and left the information exposed to anyone on the internet; no password was required. The complaint says the respondent did not assess the vendor's security measures. - **Access path:** [Public storage, databases and services](/access-paths/public-resource/); also [Third-party access](/access-paths/third-party-access/) - **Kind of control that addresses it:** assess a vendor's security before sharing sensitive documents, and block anonymous read access on cloud servers and storage that hold them ## In the words of the source “OpticsML misconfigured both the server and the storage location, leaving the sensitive personal information of tens of thousands of consumers exposed to anyone on the internet for a year, beginning in January 2018.” (regulator filing, [FTC complaint, In the Matter of Ascension Data & Analytics, LLC (Docket C-4758)](https://www.ftc.gov/system/files/documents/cases/1923126ascension_complaint.pdf)) ## Primary sources - regulator filing: [FTC complaint, In the Matter of Ascension Data & Analytics, LLC (Docket C-4758)](https://www.ftc.gov/system/files/documents/cases/1923126ascension_complaint.pdf), dated 15 Dec 2020 - regulator filing: [Mortgage Analytics Company Settles FTC Allegations It Failed to Ensure Vendor Was Adequately Protecting Personal Data](https://www.ftc.gov/news-events/news/press-releases/2020/12/mortgage-analytics-company-settles-ftc-allegations-it-failed-ensure-vendor-was-adequately-protecting), dated 15 Dec 2020 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Ascension%20Data%20%26%20Analytics%20(ascension-data-2019)) with the entry name in the subject. ## Related pages - [SolarWinds 2020: third-party access](/incidents/solarwinds-2020/) - [Chan Brothers Travel Pte Ltd 2020: pdpc](/incidents/chan-brothers-travel-pdpc-2019/) - [MRI Diagnostics Pte Ltd and Clarity Radiology Pte Ltd 2020](/incidents/mri-diagnostics-clarity-radiology-pdpc-2018/) ## Sources - FTC complaint, In the Matter of Ascension Data & Analytics, LLC (Docket C-4758): https://www.ftc.gov/system/files/documents/cases/1923126ascension_complaint.pdf (read 16 Nov 2026) - Mortgage Analytics Company Settles FTC Allegations It Failed to Ensure Vendor Was Adequately Protecting Personal Data: https://www.ftc.gov/news-events/news/press-releases/2020/12/mortgage-analytics-company-settles-ftc-allegations-it-failed-ensure-vendor-was-adequately-protecting (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/ascension-data-2019/ --- # SolarWinds (2020): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. SolarWinds disclosed the incident on 14 Dec 2020. The ledger reads the access path as third-party access with stolen user credentials and session tokens. The cloud or provider involved: Microsoft 365. Every field rests on the primary sources listed below. - **Disclosed:** 14 Dec 2020 - **Incident window:** Orion updates released between March and June 2020 (the filing's Relevant Period) - **Cloud or provider:** Microsoft 365 - **What was exposed:** The filing says a vulnerability was inserted in Orion monitoring product updates; fewer than 18,000 of approximately 33,000 notified Orion customers may have had an installation containing it. The filing also says the company's own Microsoft Office 365 email was compromised and that it had uncovered no evidence at that time of exfiltration of customer, personnel or other data. The Microsoft post describes actions in the Microsoft cloud against customer tenants. - **How access was gained:** The filing says the vulnerability was introduced as a result of a compromise of the Orion software build system, and that the company was made aware of an attack vector used to compromise its Office 365 emails. The Microsoft post says that in actions observed in the Microsoft cloud, attackers gained administrative access either using compromised privileged account credentials (for example stolen passwords) or by forging SAML tokens using compromised token signing certificates. - **Access path:** [Third-party access](/access-paths/third-party-access/); also [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** treat vendor-supplied software updates as untrusted: restrict the network reach and privileges of vendor management tools, and protect federation signing keys and administrative identities with hardware-backed storage and strong authentication ## In the words of the source “was introduced as a result of a compromise of the Orion software build system and was not present in the source code repository of the Orion products.” (company statement, [SolarWinds Form 8-K dated December 14, 2020](https://www.sec.gov/Archives/edgar/data/1739942/000162828020017451/swi-20201214.htm)) ## Primary sources - company statement: [SolarWinds Form 8-K dated December 14, 2020](https://www.sec.gov/Archives/edgar/data/1739942/000162828020017451/swi-20201214.htm), dated 14 Dec 2020 - provider post mortem: [Customer Guidance on Recent Nation-State Cyber Attacks (Microsoft Security Response Center)](https://www.microsoft.com/en-us/msrc/blog/2020/12/customer-guidance-on-recent-nation-state-cyber-attacks), dated 13 Dec 2020 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20SolarWinds%20(solarwinds-2020)) with the entry name in the subject. ## Related pages - [Chan Brothers Travel Pte Ltd 2020: pdpc](/incidents/chan-brothers-travel-pdpc-2019/) - [MRI Diagnostics Pte Ltd and Clarity Radiology Pte Ltd 2020](/incidents/mri-diagnostics-clarity-radiology-pdpc-2018/) - [Singapore Medical Association 2020: pdpc](/incidents/singapore-medical-association-pdpc-2020/) ## Sources - SolarWinds Form 8-K dated December 14, 2020: https://www.sec.gov/Archives/edgar/data/1739942/000162828020017451/swi-20201214.htm (read 16 Nov 2026) - Customer Guidance on Recent Nation-State Cyber Attacks (Microsoft Security Response Center): https://www.microsoft.com/en-us/msrc/blog/2020/12/customer-guidance-on-recent-nation-state-cyber-attacks (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/solarwinds-2020/ --- # Chan Brothers Travel Pte Ltd (2020): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Chan Brothers Travel Pte Ltd disclosed the incident on 16 Oct 2020. What was exposed, as the sources state it: The PDPC decision states the personal data of an estimated 5,593 individuals (names, email addresses and mobile numbers submitted through enquiry and feedback forms) was indexed, cached and searchable through web search engines. - **Disclosed:** 16 Oct 2020 - **Incident window:** March 2018 to May 2019 - **Cloud or provider:** Salesforce - **What was exposed:** The PDPC decision states the personal data of an estimated 5,593 individuals (names, email addresses and mobile numbers submitted through enquiry and feedback forms) was indexed, cached and searchable through web search engines. - **How access was gained:** The decision says the company hosted its site on Salesforce Community Cloud with the access configuration set to Public. A March 2018 release added an automated search engine optimisation feature, and the forms inherited the public setting, so the data was indexed. The company said it received no notice of the release; the decision found the release notes were published. - **Access path:** [Public storage, databases and services](/access-paths/public-resource/) - **Kind of control that addresses it:** review default visibility settings after each SaaS feature release, keep forms collecting personal data private, and monitor vendor release notes ## In the words of the source “As the Website’s access configuration was set to Public, the Forms automatically inherited the same setting for the purpose of the SEO feature.” (regulator filing, [PDPC decision: Chan Brothers Travel Pte Ltd](https://www.pdpc.gov.sg/assets/c4b13045-2ac1-42e8-bd76-b17f43d1ad47)) ## Primary sources - regulator filing: [PDPC decision: Chan Brothers Travel Pte Ltd](https://www.pdpc.gov.sg/assets/c4b13045-2ac1-42e8-bd76-b17f43d1ad47), dated 16 Oct 2020 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Chan%20Brothers%20Travel%20Pte%20Ltd%20(chan-brothers-travel-pdpc-2019)) with the entry name in the subject. ## Related pages - [MRI Diagnostics Pte Ltd and Clarity Radiology Pte Ltd 2020](/incidents/mri-diagnostics-clarity-radiology-pdpc-2018/) - [Singapore Medical Association 2020: pdpc](/incidents/singapore-medical-association-pdpc-2020/) - [Cisco Systems 2020: webex aws](/incidents/cisco-webex-aws-2018/) ## Sources - PDPC decision: Chan Brothers Travel Pte Ltd: https://www.pdpc.gov.sg/assets/c4b13045-2ac1-42e8-bd76-b17f43d1ad47 (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/chan-brothers-travel-pdpc-2019/ --- # MRI Diagnostics Pte Ltd and Clarity Radiology Pte Ltd (2020): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. MRI Diagnostics Pte Ltd and Clarity Radiology Pte Ltd disclosed the incident on 16 Oct 2020. What was exposed, as the sources state it: The PDPC decision states three Excel spreadsheets exposing the names, identity card numbers and type of radiology scan of approximately 4,099 individuals were accessible over the internet. - **Disclosed:** 16 Oct 2020 - **Incident window:** reported 7 November 2018 - **Cloud or provider:** not public - **What was exposed:** The PDPC decision states three Excel spreadsheets exposing the names, identity card numbers and type of radiology scan of approximately 4,099 individuals were accessible over the internet. - **How access was gained:** The decision says the clinic subscribed to an internet-based teleradiology system provided by Clarity Radiology, a cloud-based software service, which engaged an overseas vendor to maintain it. The vendor removed access restrictions on a network folder to patch the system and did not restore them, so the spreadsheets were indexed by search engines. - **Access path:** [Public storage, databases and services](/access-paths/public-resource/); also [Third-party access](/access-paths/third-party-access/) - **Kind of control that addresses it:** restore and verify access controls after maintenance, require vendors to document change and rollback steps, and scan for publicly indexed files ## In the words of the source “Without access restrictions, the Excel Spreadsheets (containing the Personal Data Sets) were indexed by Google’s search engines and exposed to unauthorised access.” (regulator filing, [PDPC decision: MRI Diagnostics Pte Ltd and Clarity Radiology Pte Ltd](https://www.pdpc.gov.sg/assets/567882ff-bcde-4d39-a654-eae013e5889b)) ## Primary sources - regulator filing: [PDPC decision: MRI Diagnostics Pte Ltd and Clarity Radiology Pte Ltd](https://www.pdpc.gov.sg/assets/567882ff-bcde-4d39-a654-eae013e5889b), dated 16 Oct 2020 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20MRI%20Diagnostics%20Pte%20Ltd%20and%20Clarity%20Radiology%20Pte%20Ltd%20(mri-diagnostics-clarity-radiology-pdpc-2018)) with the entry name in the subject. ## Related pages - [Singapore Medical Association 2020: pdpc](/incidents/singapore-medical-association-pdpc-2020/) - [Cisco Systems 2020: webex aws](/incidents/cisco-webex-aws-2018/) - [Twilio 2020: public storage, databases and services](/incidents/twilio-2020/) ## Sources - PDPC decision: MRI Diagnostics Pte Ltd and Clarity Radiology Pte Ltd: https://www.pdpc.gov.sg/assets/567882ff-bcde-4d39-a654-eae013e5889b (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/mri-diagnostics-clarity-radiology-pdpc-2018/ --- # Singapore Medical Association (2020): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Singapore Medical Association disclosed the incident on 10 Sep 2020. What was exposed, as the sources state it: The PDPC decision states the personal data of 68 individuals in 137 emails (identity card numbers, dates of birth, indemnity coverage details, education and financial transaction information) was forwarded to an external address. - **Disclosed:** 10 Sep 2020 - **Incident window:** 28 to 30 January 2020 - **Cloud or provider:** Microsoft 365 - **What was exposed:** The PDPC decision states the personal data of 68 individuals in 137 emails (identity card numbers, dates of birth, indemnity coverage details, education and financial transaction information) was forwarded to an external address. - **How access was gained:** The decision says the association believed an unauthorised user entered a Microsoft Office 365 email account by a brute force attack but had no logs to confirm, then created a rule forwarding received emails to an external address. The account password had not been changed since November 2013 and there were no measures against repeated login attempts. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** require MFA on every mailbox, lock out repeated failed sign-ins, and alert on new mail forwarding rules ## In the words of the source “The Organisation believed an unauthorised user ( UU ) gained entry into the affected Microsoft Office 365 email account by a brute force attack but did not have the system logs to confirm this.” (regulator filing, [PDPC decision: Singapore Medical Association](https://www.pdpc.gov.sg/assets/6db1d4a3-1e2f-4271-a7b6-292bfed8bfa1)) ## Primary sources - regulator filing: [PDPC decision: Singapore Medical Association](https://www.pdpc.gov.sg/assets/6db1d4a3-1e2f-4271-a7b6-292bfed8bfa1), dated 10 Sep 2020 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Singapore%20Medical%20Association%20(singapore-medical-association-pdpc-2020)) with the entry name in the subject. ## Related pages - [Cisco Systems 2020: webex aws](/incidents/cisco-webex-aws-2018/) - [Twilio 2020: public storage, databases and services](/incidents/twilio-2020/) - [Twitter 2020: stolen user credentials and session tokens](/incidents/twitter-2020/) ## Sources - PDPC decision: Singapore Medical Association: https://www.pdpc.gov.sg/assets/6db1d4a3-1e2f-4271-a7b6-292bfed8bfa1 (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/singapore-medical-association-pdpc-2020/ --- # Cisco Systems (2020): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Cisco Systems disclosed the incident on 26 Aug 2020. The ledger reads the access path as retained access after departure or change. The cloud or provider involved: AWS. Every field rests on the primary sources listed below. Fields no primary source states read not public. - **Disclosed:** 26 Aug 2020 - **Incident window:** September 24, 2018 - **Cloud or provider:** AWS - **What was exposed:** The press release says 456 virtual machines for the WebEx Teams application were deleted and over 16,000 WebEx Teams accounts were shut down for up to two weeks. It says the company spent approximately $1,400,000 in employee time and refunded over $1,000,000 to affected customers, and that no customer data was compromised. - **How access was gained:** The Department of Justice says that under the plea agreement an individual who had worked for Cisco and resigned in approximately April 2018 admitted intentionally accessing Cisco's cloud infrastructure hosted by Amazon Web Services without permission on September 24, 2018, and deployed code from a Google Cloud Project account that deleted the virtual machines. - **Access path:** [Retained access after departure or change](/access-paths/retained-access/) - **Kind of control that addresses it:** remove cloud access for departing staff on their last day and regularly review identities, keys and tokens that belong to former employees and contractors ## In the words of the source “intentionally accessing Cisco Systems’ cloud infrastructure that was hosted by Amazon Web Services without Cisco’s permission on September 24, 2018.” (court document, [San Jose Man Pleads Guilty To Damaging Cisco's Network (U.S. Attorney's Office, Northern District of California)](https://www.justice.gov/usao-ndca/pr/san-jose-man-pleads-guilty-damaging-cisco-s-network)) ## Primary sources - court document: [San Jose Man Pleads Guilty To Damaging Cisco's Network (U.S. Attorney's Office, Northern District of California)](https://www.justice.gov/usao-ndca/pr/san-jose-man-pleads-guilty-damaging-cisco-s-network), dated 26 Aug 2020 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Cisco%20Systems%20(cisco-webex-aws-2018)) with the entry name in the subject. ## Related pages - [Twilio 2020: public storage, databases and services](/incidents/twilio-2020/) - [Twitter 2020: stolen user credentials and session tokens](/incidents/twitter-2020/) - [Blackbaud Inc. 2020: over-permissioned identities](/incidents/blackbaud-2020/) ## Sources - San Jose Man Pleads Guilty To Damaging Cisco's Network (U.S. Attorney's Office, Northern District of California): https://www.justice.gov/usao-ndca/pr/san-jose-man-pleads-guilty-damaging-cisco-s-network (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/cisco-webex-aws-2018/ --- # Twilio (2020): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Twilio disclosed the incident on 22 Jul 2020. The ledger reads the access path as public storage, databases and services. The cloud or provider involved: AWS. Every field rests on the primary sources listed below. Fields no primary source states read not public. - **Disclosed:** 22 Jul 2020 - **Incident window:** July 19, 2020 1:12 PM PDT to July 20, 2020 10:30 PM PDT (window for downloads of the modified library, per the company) - **Cloud or provider:** AWS - **What was exposed:** A modified version (v1.20) of the TaskRouter JS SDK hosted on the company's content domain. The company stated it had no evidence that any customer data was accessed and that no malicious party had access to its internal systems, code or data; customers who downloaded v1.20 in the window were asked to replace it. - **How access was gained:** The company's incident report says a misconfiguration in the S3 bucket hosting the library let a bad actor inject code, and that the access policy for the path storing the SDK allowed anybody on the Internet to read and write to that path. - **Access path:** [Public storage, databases and services](/access-paths/public-resource/) - **Kind of control that addresses it:** deny public write access on storage that serves code to customers, audit bucket policies on a schedule, and publish integrity checks for served scripts ## In the words of the source “We had not properly configured the access policy for one of our AWS S3 buckets.” (company statement, [Incident Report: TaskRouter JS SDK Security Incident - July 19, 2020](https://www.twilio.com/en-us/blog/company/communications/incident-report-taskrouter-js-sdk-july-2020)) ## Primary sources - company statement: [Incident Report: TaskRouter JS SDK Security Incident - July 19, 2020](https://www.twilio.com/en-us/blog/company/communications/incident-report-taskrouter-js-sdk-july-2020), dated 22 Jul 2020 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Twilio%20(twilio-2020)) with the entry name in the subject. ## Related pages - [Twitter 2020: stolen user credentials and session tokens](/incidents/twitter-2020/) - [Blackbaud Inc. 2020: over-permissioned identities](/incidents/blackbaud-2020/) - [LiveAuctioneers 2020: third-party access](/incidents/liveauctioneers-2020/) ## Sources - Incident Report: TaskRouter JS SDK Security Incident - July 19, 2020: https://www.twilio.com/en-us/blog/company/communications/incident-report-taskrouter-js-sdk-july-2020 (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/twilio-2020/ --- # Twitter (2020): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Twitter disclosed the incident on 18 Jul 2020. What was exposed, as the sources state it: The company stated attackers targeted 130 Twitter accounts, tweeted from 45, accessed the direct message inbox of 36 and downloaded the Twitter Data of 7. - **Disclosed:** 18 Jul 2020 - **Incident window:** July 15, 2020 - **Cloud or provider:** Twitter - **What was exposed:** The company stated attackers targeted 130 Twitter accounts, tweeted from 45, accessed the direct message inbox of 36 and downloaded the Twitter Data of 7. - **How access was gained:** The company says a phone spear phishing attack targeted a small number of employees; the attackers used the employees' credentials to access internal systems, including getting through two-factor protections, then used the credentials of employees with access to account support tools. - **Access path:** [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) - **Kind of control that addresses it:** require phishing-resistant multi-factor authentication for every identity with access to internal support tools, with a second approver for account-changing actions ## In the words of the source “The attackers successfully manipulated a small number of employees and used their credentials to access Twitter’s internal systems, including getting through our two-factor protections.” (company statement, [An update on our security incident (archived copy of the company blog, updated July 30, 2020)](https://web.archive.org/web/20200801/https://blog.twitter.com/en_us/topics/company/2020/an-update-on-our-security-incident)) ## Primary sources - company statement: [An update on our security incident (archived copy of the company blog, updated July 30, 2020)](https://web.archive.org/web/20200801/https://blog.twitter.com/en_us/topics/company/2020/an-update-on-our-security-incident), dated 30 Jul 2020 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Twitter%20(twitter-2020)) with the entry name in the subject. ## Related pages - [Blackbaud Inc. 2020: over-permissioned identities](/incidents/blackbaud-2020/) - [LiveAuctioneers 2020: third-party access](/incidents/liveauctioneers-2020/) - [Microsoft 2020: support database](/incidents/microsoft-support-database-2019/) ## Sources - An update on our security incident (archived copy of the company blog, updated July 30, 2020): https://web.archive.org/web/20200801/https://blog.twitter.com/en_us/topics/company/2020/an-update-on-our-security-incident (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/twitter-2020/ --- # Blackbaud Inc. (2020): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Blackbaud Inc. disclosed the incident on 16 Jul 2020. The ledger reads the access path as over-permissioned identities. The cloud or provider involved: Blackbaud (software provider; the New York Attorney General calls it a cloud company). Every field rests on the primary sources listed below. - **Disclosed:** 16 Jul 2020 - **Incident window:** Early 2020; the FTC says the intrusion went undetected for three months - **Cloud or provider:** Blackbaud (software provider; the New York Attorney General calls it a cloud company) - **What was exposed:** The SEC states the attacker accessed and exfiltrated donor bank account information and Social Security numbers, and that more than 13,000 customers were impacted. The FTC states the personal data of millions of consumers, including Social Security and bank account numbers, was removed unencrypted. The New York Attorney General states the breach exposed personal information of customers and millions of their donors and constituents. - **How access was gained:** The FTC press release says that, according to the complaint, a hacker in early 2020 accessed a customer’s Blackbaud-hosted database and, once logged in, moved across multiple Blackbaud-hosted environments by using existing vulnerabilities and local administrator accounts and by creating new administrator accounts. The FTC says the company failed to segment data, adequately implement multifactor authentication and monitor for breach attempts. The company paid a ransom of 24 Bitcoin, according to the FTC. - **Access path:** [Over-permissioned identities](/access-paths/over-permissioned-identity/) - **Kind of control that addresses it:** segment hosted customer environments from one another, limit local administrator accounts, require multifactor authentication on every identity with access to hosted data, and monitor administrator account creation ## In the words of the source “As a result of these failures, a hacker in early 2020 accessed a customer’s Blackbaud-hosted database, according to the complaint.” (regulator filing, [FTC press release: FTC Order Will Require Blackbaud to Delete Unnecessary Data, Boost Safeguards to Settle Charges its Lax Security Practices Led to Data Breach (web.archive.org copy)](https://web.archive.org/web/2024/https://www.ftc.gov/news-events/news/press-releases/2024/02/ftc-order-will-require-blackbaud-delete-unnecessary-data-boost-safeguards-settle-charges-its-lax)) ## Primary sources - regulator filing: [FTC press release: FTC Order Will Require Blackbaud to Delete Unnecessary Data, Boost Safeguards to Settle Charges its Lax Security Practices Led to Data Breach (web.archive.org copy)](https://web.archive.org/web/2024/https://www.ftc.gov/news-events/news/press-releases/2024/02/ftc-order-will-require-blackbaud-delete-unnecessary-data-boost-safeguards-settle-charges-its-lax), dated 1 Feb 2024 - regulator filing: [SEC press release 2023-48: SEC Charges Software Company Blackbaud Inc. for Misleading Disclosures About Ransomware Attack That Impacted Charitable Donors](https://www.sec.gov/news/press-release/2023-48), dated 9 Mar 2023 - regulator filing: [New York Attorney General press release: Attorney General James and Multistate Coalition Secure $49.5 Million from Cloud Company for Data Breach](https://ag.ny.gov/press-release/2023/attorney-general-james-and-multistate-coalition-secure-495-million-cloud-company), dated 5 Oct 2023 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Blackbaud%20Inc.%20(blackbaud-2020)) with the entry name in the subject. ## Related pages - [LiveAuctioneers 2020: third-party access](/incidents/liveauctioneers-2020/) - [Microsoft 2020: support database](/incidents/microsoft-support-database-2019/) - [Imperva 2019: leaked access keys and secrets](/incidents/imperva-2019/) ## Sources - FTC press release: FTC Order Will Require Blackbaud to Delete Unnecessary Data, Boost Safeguards to Settle Charges its Lax Security Practices Led to Data Breach (web.archive.org copy): https://web.archive.org/web/2024/https://www.ftc.gov/news-events/news/press-releases/2024/02/ftc-order-will-require-blackbaud-delete-unnecessary-data-boost-safeguards-settle-charges-its-lax (read 16 Nov 2026) - SEC press release 2023-48: SEC Charges Software Company Blackbaud Inc. for Misleading Disclosures About Ransomware Attack That Impacted Charitable Donors: https://www.sec.gov/news/press-release/2023-48 (read 16 Nov 2026) - New York Attorney General press release: Attorney General James and Multistate Coalition Secure $49.5 Million from Cloud Company for Data Breach: https://ag.ny.gov/press-release/2023/attorney-general-james-and-multistate-coalition-secure-495-million-cloud-company (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/blackbaud-2020/ --- # LiveAuctioneers (2020): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. LiveAuctioneers disclosed the incident on 11 Jul 2020. The ledger reads the access path as third-party access with leaked access keys and secrets. The cloud or provider involved: AWS. Every field rests on the primary sources listed below. Fields no primary source states read not public. - **Disclosed:** 11 Jul 2020 - **Incident window:** June 19, 2020 to July 11, 2020 (Washington Attorney General record) - **Cloud or provider:** AWS - **What was exposed:** The company's letter says 38,523 Washington residents were involved; the user database held names, email addresses (user names), MD5-hashed passwords, contact details, IP addresses used to create accounts and records of past sales transactions. It says no Social Security numbers or financial data were collected. - **How access was gained:** The company's letter to the Washington Attorney General says a hacker breached one of its third-party software solutions and obtained a source control key, which gave access to its source repository, its cloud-based storage environment and its production database, then used internal users' credentials to download the user database. - **Access path:** [Third-party access](/access-paths/third-party-access/); also [Leaked access keys and secrets](/access-paths/leaked-access-key/) - **Kind of control that addresses it:** scope and rotate source-control keys issued to third-party tools, keep production credentials out of repositories, and alert on bulk database reads ## In the words of the source “the Hacker obtained a source control key to gain access to our source controller repository maintained by another third party software developed, our cloud-based storage environment, and our production database.” (company statement, [LiveAuctioneers letter to the Washington Attorney General, filed with the state breach notification portal](https://agportal-s3bucket.s3.amazonaws.com/databreach/BreachM10283.pdf)) ## Primary sources - company statement: [LiveAuctioneers letter to the Washington Attorney General, filed with the state breach notification portal](https://agportal-s3bucket.s3.amazonaws.com/databreach/BreachM10283.pdf), dated 3 Sep 2020 - regulator filing: [Washington Attorney General data breach record: Live Auctioneers](https://www.atg.wa.gov/live-auctioneers), dated 3 Sep 2020 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20LiveAuctioneers%20(liveauctioneers-2020)) with the entry name in the subject. ## Related pages - [Microsoft 2020: support database](/incidents/microsoft-support-database-2019/) - [Imperva 2019: leaked access keys and secrets](/incidents/imperva-2019/) - [Capital One 2019: exposed network services and request pat](/incidents/capital-one-2019/) ## Sources - LiveAuctioneers letter to the Washington Attorney General, filed with the state breach notification portal: https://agportal-s3bucket.s3.amazonaws.com/databreach/BreachM10283.pdf (read 16 Nov 2026) - Washington Attorney General data breach record: Live Auctioneers: https://www.atg.wa.gov/live-auctioneers (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/liveauctioneers-2020/ --- # Microsoft (2020): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Microsoft disclosed the incident on 22 Jan 2020. The ledger reads the access path as public storage, databases and services. The cloud or provider involved: Microsoft Azure. Every field rests on the primary sources listed below. Fields no primary source states read not public. - **Disclosed:** 22 Jan 2020 - **Incident window:** December 5, 2019 (network security group change) to December 31, 2019 (remediation) - **Cloud or provider:** Microsoft Azure - **What was exposed:** An internal customer support database used for support case analytics. The post says most customers did not have personally identifiable information exposed, that records are redacted with automated tools, and that some data may have remained unredacted, for example an email address written in a non-standard format. Counts: not public. - **How access was gained:** The post says a change made to the database's network security group on December 5, 2019 contained misconfigured security rules that enabled exposure of the data. It says the investigation found no malicious use, and that the issue was specific to an internal database and does not represent an exposure of the company's commercial cloud services. - **Access path:** [Public storage, databases and services](/access-paths/public-resource/) - **Kind of control that addresses it:** continuously check network security rules on data stores for rules that allow unrestricted inbound access, and alert on rule changes ## In the words of the source “Our investigation has determined that a change made to the database’s network security group on December 5, 2019 contained misconfigured security rules that enabled exposure of the data.” (provider post mortem, [Access Misconfiguration for Customer Support Database (Microsoft Security Response Center)](https://www.microsoft.com/en-us/msrc/blog/2020/01/access-misconfiguration-for-customer-support-database)) ## Primary sources - provider post mortem: [Access Misconfiguration for Customer Support Database (Microsoft Security Response Center)](https://www.microsoft.com/en-us/msrc/blog/2020/01/access-misconfiguration-for-customer-support-database), dated 22 Jan 2020 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Microsoft%20(microsoft-support-database-2019)) with the entry name in the subject. ## Related pages - [Imperva 2019: leaked access keys and secrets](/incidents/imperva-2019/) - [Capital One 2019: exposed network services and request pat](/incidents/capital-one-2019/) - [Online Buddies (Jack’d dating app) 2019: jackd s3](/incidents/online-buddies-jackd-s3-2018/) ## Sources - Access Misconfiguration for Customer Support Database (Microsoft Security Response Center): https://www.microsoft.com/en-us/msrc/blog/2020/01/access-misconfiguration-for-customer-support-database (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/microsoft-support-database-2019/ --- # Imperva (2019): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Imperva disclosed the incident on 27 Aug 2019. What was exposed, as the sources state it: Elements of the Incapsula (Cloud WAF) customer database through September 15, 2017: email addresses and hashed and salted passwords, and for a subset of customers API keys and customer-provided SSL certificates. The company stated the exposure came from a database snapshot. - **Disclosed:** 27 Aug 2019 - **Incident window:** October 2018 (unauthorized use of an administrative API key, per the company); the company received the data set on August 20, 2019 - **Cloud or provider:** AWS - **What was exposed:** Elements of the Incapsula (Cloud WAF) customer database through September 15, 2017: email addresses and hashed and salted passwords, and for a subset of customers API keys and customer-provided SSL certificates. The company stated the exposure came from a database snapshot. - **How access was gained:** The company's follow-up post says an unauthorized use of an administrative API key in one of its production AWS accounts led to exposure of a database snapshot. It lists the steps as a snapshot created for testing, an internal compute instance accessible from the outside world that contained an AWS API key, compromise of that instance and theft of the key, and use of the key to access the snapshot. - **Access path:** [Leaked access keys and secrets](/access-paths/leaked-access-key/); also [Exposed network services and request paths](/access-paths/exposed-network-service/) - **Kind of control that addresses it:** keep long-lived cloud administrative keys out of compute instances, place internal instances on a private network by default, and expire or rotate keys on a schedule ## In the words of the source “Our investigation identified an unauthorized use of an administrative API key in one of our production AWS accounts in October 2018, which led to an exposure of a database snapshot containing emails and hashed & salted passwords.” (company statement, [Imperva Security Update (follow-up on the root cause), archived copy of the company blog](https://web.archive.org/web/20191101/https://www.imperva.com/blog/ceoblog/)) ## Primary sources - company statement: [Imperva Security Update (follow-up on the root cause), archived copy of the company blog](https://web.archive.org/web/20191101/https://www.imperva.com/blog/ceoblog/), dated 10 Oct 2019 - company statement: [Imperva Security Update (initial announcement), archived copy of the company blog](https://web.archive.org/web/20190901/https://www.imperva.com/blog/ceoblog/), dated 27 Aug 2019 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Imperva%20(imperva-2019)) with the entry name in the subject. ## Related pages - [Capital One 2019: exposed network services and request pat](/incidents/capital-one-2019/) - [Online Buddies (Jack’d dating app) 2019: jackd s3](/incidents/online-buddies-jackd-s3-2018/) - [Veradigm Inc. 2026: vendor api](/incidents/veradigm-vendor-api-2026/) ## Sources - Imperva Security Update (follow-up on the root cause), archived copy of the company blog: https://web.archive.org/web/20191101/https://www.imperva.com/blog/ceoblog/ (read 16 Nov 2026) - Imperva Security Update (initial announcement), archived copy of the company blog: https://web.archive.org/web/20190901/https://www.imperva.com/blog/ceoblog/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/imperva-2019/ --- # Capital One (2019): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Capital One disclosed the incident on 29 Jul 2019. The ledger reads the access path as exposed network services and request paths. The cloud or provider involved: AWS. Every field rests on the primary sources listed below. Fields no primary source states read not public. - **Disclosed:** 29 Jul 2019 - **Incident window:** March 22 and 23, 2019 (the company's 8-K); the company determined the incident on July 19, 2019 - **Cloud or provider:** AWS - **What was exposed:** The company stated the event affected approximately 100 million individuals in the United States and approximately 6 million in Canada. Credit card application data from 2005 through early 2019 (names, addresses, ZIP or postal codes, phone numbers, email addresses, dates of birth, self-reported income), portions of credit card customer data, about 140,000 Social Security numbers and about 80,000 linked bank account numbers of US secured credit card customers, and approximately 1 million Canadian Social Insurance Numbers. The company stated no credit card account numbers or log-in credentials were compromised. - **How access was gained:** The company stated that a highly sophisticated individual was able to exploit a specific configuration vulnerability in its infrastructure and that the issue was not specific to the cloud. The OCC consent order finds the bank failed to establish appropriate risk management for its cloud operating environment, including network security controls. - **Access path:** [Exposed network services and request paths](/access-paths/exposed-network-service/) - **Kind of control that addresses it:** review and test the configuration of internet-facing request-handling components, and block workloads from reaching cloud credential services they do not need ## In the words of the source “a highly sophisticated individual was able to exploit a specific configuration vulnerability in our infrastructure.” (company statement, [Capital One Form 8-K dated July 29, 2019, Exhibit 99.1 press release and questions](https://www.sec.gov/Archives/edgar/data/927628/000092762819000262/exhibit991-pressrelease72919.htm)) ## Primary sources - company statement: [Capital One Form 8-K dated July 29, 2019, Exhibit 99.1 press release and questions](https://www.sec.gov/Archives/edgar/data/927628/000092762819000262/exhibit991-pressrelease72919.htm), dated 29 Jul 2019 - company statement: [Information on the Capital One cyber incident](https://www.capitalone.com/digital/facts2019/), dated not stated - regulator filing: [OCC consent order, Capital One, N.A. (AA-EC-20-51)](https://www.occ.gov/static/enforcement-actions/ea2020-036.pdf), dated 5 Aug 2020 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Capital%20One%20(capital-one-2019)) with the entry name in the subject. ## Related pages - [Online Buddies (Jack’d dating app) 2019: jackd s3](/incidents/online-buddies-jackd-s3-2018/) - [Veradigm Inc. 2026: vendor api](/incidents/veradigm-vendor-api-2026/) - [IEH Corporation 2026: microsoft 365](/incidents/ieh-microsoft-365-2026/) ## Sources - Capital One Form 8-K dated July 29, 2019, Exhibit 99.1 press release and questions: https://www.sec.gov/Archives/edgar/data/927628/000092762819000262/exhibit991-pressrelease72919.htm (read 16 Nov 2026) - Information on the Capital One cyber incident: https://www.capitalone.com/digital/facts2019/ (read 16 Nov 2026) - OCC consent order, Capital One, N.A. (AA-EC-20-51): https://www.occ.gov/static/enforcement-actions/ea2020-036.pdf (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/capital-one-2019/ --- # Online Buddies (Jack’d dating app) (2019): what the sources say about the cloud incident Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Online Buddies (Jack’d dating app) disclosed the incident on 28 Jun 2019. What was exposed, as the sources state it: The release says private photos (which may have included nude images), public photos and personal information (location, device ID, last login date, hashed password) were at risk of unauthorized access; counts not public. - **Disclosed:** 28 Jun 2019 - **Incident window:** vulnerability reported to management in February 2018, as the release says - **Cloud or provider:** AWS - **What was exposed:** The release says private photos (which may have included nude images), public photos and personal information (location, device ID, last login date, hashed password) were at risk of unauthorized access; counts not public. - **How access was gained:** The New York Attorney General’s investigation confirmed the company failed to secure user photos stored in an AWS S3 storage service, and left app interfaces to backend data unsecured. - **Access path:** [Public storage, databases and services](/access-paths/public-resource/) - **Kind of control that addresses it:** block public access to object storage by default, require signed access for user media and test app interfaces to backend data for authorization ## In the words of the source “Online Buddies failed to secure data — including users’ private photos — that the company had stored using Amazon Web Services Simple Storage Service (S3).” (regulator filing, [NY Attorney General: Attorney General James Announces Settlement With Dating App For Failure To Secure Private And Nude Photographs](https://ag.ny.gov/press-release/2019/attorney-general-james-announces-settlement-dating-app-failure-secure-private-and)) ## Primary sources - regulator filing: [NY Attorney General: Attorney General James Announces Settlement With Dating App For Failure To Secure Private And Nude Photographs](https://ag.ny.gov/press-release/2019/attorney-general-james-announces-settlement-dating-app-failure-secure-private-and), dated 28 Jun 2019 If a field is wrong or a newer primary source exists, [write to the editor](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Online%20Buddies%20(Jack%E2%80%99d%20dating%20app)%20(online-buddies-jackd-s3-2018)) with the entry name in the subject. ## Related pages - [Veradigm Inc. 2026: vendor api](/incidents/veradigm-vendor-api-2026/) - [IEH Corporation 2026: microsoft 365](/incidents/ieh-microsoft-365-2026/) - [Amgen 2026: cloud environments](/incidents/amgen-cloud-environments-2026/) ## Sources - NY Attorney General: Attorney General James Announces Settlement With Dating App For Failure To Secure Private And Nude Photographs: https://ag.ny.gov/press-release/2019/attorney-general-james-announces-settlement-dating-app-failure-secure-private-and (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/incidents/online-buddies-jackd-s3-2018/ --- # The eight access paths, counted across the ledger Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Eight access paths explain how access to cloud data was gained in the ledger: leaked access keys, stolen user credentials, over-permissioned identities, public resources, cross-tenant flaws, third-party access, exposed network services and retained access. Each path has a definition, a two-paragraph explanation from provider documentation and the kind of preventive control. | Access path | Incidents | Kind of control | | --- | --- | --- | | [Leaked access keys and secrets](/access-paths/leaked-access-key/) | 27 | replace long-lived keys with short-lived credentials issued to the workload, and scan code, logs and files for secrets before they are published | | [Stolen user credentials and session tokens](/access-paths/stolen-user-credentials/) | 47 | require phishing-resistant multifactor authentication for every person with access to the data platform, and limit what any single sign-in can reach | | [Over-permissioned identities](/access-paths/over-permissioned-identity/) | 13 | scope each identity to the actions and resources its task needs, cap delegated permissions with boundaries, and remove access that is unused | | [Public storage, databases and services](/access-paths/public-resource/) | 15 | enforce an account-wide or organization-wide block on public access to storage, and require sign-in for every database and service | | [Cross-tenant access](/access-paths/cross-tenant/) | 3 | validate the tenant and issuer on every token an application accepts, and turn off cross-tenant features that are not in use | | [Third-party access](/access-paths/third-party-access/) | 25 | grant each vendor a scoped role bound to a customer-specific condition, review integrations on a schedule, and revoke tokens the vendor no longer needs | | [Exposed network services and request paths](/access-paths/exposed-network-service/) | 8 | require session-based metadata access on every instance, close management ports to the internet, and limit the role attached to each workload | | [Retained access after departure or change](/access-paths/retained-access/) | 7 | remove every credential and grant tied to a departing person or retired app on the day it ends, and review unused access on a fixed schedule | ## Related pages - [Cloud breaches by cloud and provider](/clouds/) - [Cloud breaches by year, 2019 to 2026](/years/) - [Disclosed cloud attack techniques, with sources](/techniques/) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/access-paths/ --- # Leaked access keys and secrets: 27 incidents in the ledger Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. A leaked access key is a long-lived key, token or secret that ended up somewhere it should not be, such as a code repository, a container image, a log or a public file, and was then used by someone other than its owner to call a cloud provider's interfaces or reach a service. ## How it works Cloud providers issue long-lived credentials for programmatic access: a key pair for a user, a key file for a service account, or a shared access signature for a storage account. AWS documentation says access keys are long-term credentials that sign requests to its API and that its secret half can be retrieved only when the key is created. Microsoft's documentation says a leaked shared access signature can be used by anyone who obtains it. Google Cloud's documentation lists credential leakage as a risk of service account keys. The credential carries no information about who holds it. Whoever presents a valid key is treated as its owner, from any network, until the key is deleted or expires. For that reason the providers' guidance, read 16 Nov 2026, favors temporary credentials that expire on their own and keeping keys out of application files, and Microsoft's guidance adds short expiry times and a revocation plan for tokens that must exist. ## The incidents | Disclosed | Organization | Cloud | How access was gained | | --- | --- | --- | --- | | 2025-09-22 | [GitHub (npm registry)](/incidents/github-npm-shai-hulud-2025/) | GitHub | GitHub stated that a self-replicating worm spread through compromised maintainer accounts by injecting malicious post-install scripts into popular packages. | | 2025-09-02 | [Cloudflare](/incidents/cloudflare-salesforce-drift-2025/) | Salesforce | The company stated the threat actor obtained OAuth credentials associated with the Salesloft Drift chat agent's Salesforce integration, and used a stolen credential used by the Salesloft integration to log in to its Salesforce tenant. | | 2025-08-26 | [Nx (Nrwl)](/incidents/nx-github-actions-2025/) | GitHub | The post-mortem says attackers exploited a GitHub Actions injection vulnerability to steal the npm publishing token; the repository's workflow permissions were read and write. | | 2025-08-20 | [Salesloft (Drift)](/incidents/salesloft-drift-2025/) | AWS | The investigation summary says the threat actor accessed the Salesloft GitHub account (March to June 2025), then accessed Drift's AWS environment and obtained OAuth tokens for customers' integrations, and used them to access data via Drift integrations. | | 2025-07-23 | [Amazon Web Services](/incidents/aws-amazon-q-extension-2025/) | AWS | The bulletin says the extension had an inappropriately scoped GitHub token in its CodeBuild configuration, and with that token the threat actor committed malicious code. | | 2024-12-08 | [BeyondTrust](/incidents/beyondtrust-remote-support-saas-2024/) | Amazon Web Services | The company stated a zero-day vulnerability of a third-party application was used to gain access to an online asset in a BeyondTrust AWS account, which allowed the threat actor to obtain an infrastructure API key that was then used against a separate AWS account that operated Remote Support infrastructure. | | 2024-05-01 | [Dropbox](/incidents/dropbox-sign-2024/) | not public | The company stated a third party gained access to an automated system configuration tool, compromised a back-end service account with privileges across the production environment, and reached the customer database; it answered how access was gained with a compromised access token. | | 2024-02-01 | [Cloudflare](/incidents/cloudflare-thanksgiving-2023/) | Okta | The company stated the actor used one access token and three service account credentials taken in the October 2023 Okta compromise that the company had failed to rotate. | | 2023-11-07 | [Sumo Logic](/incidents/sumo-logic-aws-credential-2023/) | AWS | The customer notice says the activity used a compromised credential to access a Sumo Logic AWS account. How the credential was compromised is not public. | | 2023-09-18 | [Microsoft](/incidents/microsoft-ai-sas-token-2023/) | Microsoft Azure | Microsoft stated an employee shared a URL for a blob store in a public GitHub repository, and the URL included an overly-permissive Shared Access Signature token for an internal storage account. | | 2023-08-30 | [Sourcegraph](/incidents/sourcegraph-2023/) | not public | The company stated a malicious actor used a leaked site-admin access token, accidentally committed in a pull request, to raise an account to site-admin on Sourcegraph.com. | | 2023-08-16 | [E-Commerce Enablers Pte. Ltd.](/incidents/ecommerce-enablers-pdpc-2020/) | AWS | The decision says an AWS access key with full administrative privileges was committed to a private GitHub repository on 4 June 2019, removed on 6 June but still viewable in the commit history, and not fully disabled when a replacement key was created. On 9 September 2020 a threat actor used it to reach the AWS environment, changed security settings to allow remote internet access to database instances and exfiltrated data. The company conjectured the key was found in the commit history. | | 2023-07-11 | [Microsoft](/incidents/microsoft-storm-0558-2023/) | Microsoft 365 | Microsoft stated the actor used an acquired Microsoft account consumer signing key to forge tokens, and exploited a token validation issue so that enterprise mail accepted them. The post-mortem says the most probable mechanism for the key acquisition was a compromised engineer corporate account with access to a debugging environment holding a crash dump that contained the key. | | 2023-03-23 | [GitHub](/incidents/github-rsa-ssh-key-2023/) | GitHub | The company stated the exposure resulted from what it believes was an inadvertent publishing of private information. | | 2022-12-22 | [LastPass](/incidents/lastpass-cloud-storage-2022/) | not public | The company stated that an unknown threat actor accessed a cloud-based storage environment using information obtained from the August 2022 incident, and that credentials and keys obtained from another targeted employee were used to access and decrypt some storage volumes. | | 2022-12-19 | [RedMart Limited](/incidents/redmart-pdpc-2020/) | AWS and Alibaba Cloud | The decision says a threat actor gained unauthorised access to the GitHub account of a member of the DevOps team, found a high-privilege key to the AWS environment in the repositories, used it to find a second key in a private S3 bucket, created an EC2 instance, and crossed a VPN link to the Alibaba cloud database. The database was neither encrypted nor password protected. | | 2022-11-01 | [Dropbox](/incidents/dropbox-github-phishing-2022/) | GitHub | The company stated that employees received phishing emails impersonating a CI service, and that a fake login page collected GitHub credentials and a one-time code, giving the threat actor access to one GitHub organization. | | 2022-10-24 | [Drizly](/incidents/drizly-ftc-2022/) | AWS | The FTC complaint stated that in early July 2020 a malicious actor reused credentials from an unrelated breach to reach an executive's GitHub account, found AWS and database credentials stored in the company's GitHub repositories, and used them to change the company's AWS security settings and reach the production environment. | | 2022-09-15 | [MyRepublic Limited](/incidents/myrepublic-pdpc-2021/) | AWS | The decision says the Mobile Order Portal stored KYC documents in a bucket on AWS cloud storage, and the bucket's access key sat in the portal's source code. The external actor used the key to access the bucket. The company could not determine with certainty how the key was obtained but found it was disclosed in a portal page showing technical information and in a source repository open to all its developers. | | 2022-04-21 | [GeniusU Pte. Ltd.](/incidents/geniusu-pdpc-2021/) | GitHub | The decision says the likely cause was a compromise of a developer's GitHub password, which let the threat actor into the company's GitHub environment; database login credentials had been stored in the codebase there, so the actor reached and exfiltrated the database. The staging database held real personal data. | | 2022-04-15 | [Heroku (Salesforce) and Travis CI OAuth tokens](/incidents/heroku-travisci-oauth-2022/) | GitHub | Heroku stated that initial access to its environment was obtained by using a compromised token for a Heroku machine account, after which stored customer GitHub integration OAuth tokens were downloaded; GitHub stated the tokens were then abused against GitHub.com. | | 2021-11-11 | [Commeasure Pte Ltd (operates the RedDoorz booking platform)](/incidents/commeasure-reddoorz-pdpc-2020/) | AWS | The decision says the threat actors most likely obtained an AWS access key embedded in an Android application package, created in 2015, last updated in January 2018, still downloadable from the Google Play Store, and marked as a test key although it could reach a live production database in Amazon RDS. | | 2021-09-21 | [Sendtech Pte. Ltd.](/incidents/sendtech-pdpc-2021/) | AWS | The decision says the cause was a compromised AWS access key created in 2015 and not rotated since. With it the attacker gained admin privileges, created another admin account and queried the buckets storing personal data. The company suspected the key could have been compromised through former or current employees but the decision does not confirm how. | | 2021-04-15 | [Codecov](/incidents/codecov-2021/) | Codecov | The company stated that an error in its Docker image creation process allowed the actor to extract the credential required to modify the Bash Uploader script, which was then altered to send CI environment variables to a third-party server. | | 2021-03-16 | [Mimecast](/incidents/mimecast-solarwinds-certificate-2021/) | Microsoft 365 | The company stated the threat actor used the SolarWinds supply-chain compromise to gain access to part of its production grid environment, and from there accessed Mimecast-issued certificates and related connection information; the certificate was then used against a small number of customer Microsoft 365 tenants. | | 2020-07-11 | [LiveAuctioneers](/incidents/liveauctioneers-2020/) | AWS | The company's letter to the Washington Attorney General says a hacker breached one of its third-party software solutions and obtained a source control key, which gave access to its source repository, its cloud-based storage environment and its production database, then used internal users' credentials to download the user database. | | 2019-08-27 | [Imperva](/incidents/imperva-2019/) | AWS | The company's follow-up post says an unauthorized use of an administrative API key in one of its production AWS accounts led to exposure of a database snapshot. It lists the steps as a snapshot created for testing, an internal compute instance accessible from the outside world that contained an AWS API key, compromise of that instance and theft of the key, and use of the key to access the snapshot. | ## The kind of control that addresses it replace long-lived keys with short-lived credentials issued to the workload, and scan code, logs and files for secrets before they are published ## Related pages - [Stolen user credentials and session tokens: cloud breaches](/access-paths/stolen-user-credentials/) - [Over-permissioned identities: cloud breaches by access pat](/access-paths/over-permissioned-identity/) - [Public storage, databases and services: cloud breaches by ](/access-paths/public-resource/) ## Sources - AWS IAM User Guide: manage access keys for IAM users: https://docs.aws.amazon.com/IAM/latest/UserGuide/id_credentials_access-keys.html (read 16 Nov 2026) - AWS IAM User Guide: temporary security credentials: https://docs.aws.amazon.com/IAM/latest/UserGuide/id_credentials_temp.html (read 16 Nov 2026) - Microsoft Learn: shared access signatures overview: https://learn.microsoft.com/en-us/azure/storage/common/storage-sas-overview (read 16 Nov 2026) - Google Cloud: best practices for managing service account keys: https://cloud.google.com/iam/docs/best-practices-for-managing-service-account-keys (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/access-paths/leaked-access-key/ --- # Stolen user credentials and session tokens: 47 incidents in the ledger Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Stolen user credentials are the password or session token of a person, obtained by phishing, an infostealer, password spraying or a similar method, and then used to sign in as that person to a cloud console, a software-as-a-service tenant or a data platform, often where no second factor was required. ## How it works A person signs in with something they know, and a cloud or SaaS provider treats a valid sign-in as that person. Microsoft's documentation says that if only a password is used to authenticate a user, it leaves an insecure vector for attack, because a weak or exposed password can be used by an attacker. A second factor raises the cost, but the providers distinguish between factors: AWS documentation recommends phishing-resistant passkeys and security keys because they resist phishing, man-in-the-middle and replay attacks. Once a sign-in succeeds, the session is what the provider trusts. A stolen session token or an approved push prompt gives the same access as the password did, which is why the ledger records the factor type where the sources state it. The preventive layer is therefore the strength of the sign-in method and the scope of what a signed-in person can reach, not the password alone. ## The incidents | Disclosed | Organization | Cloud | How access was gained | | --- | --- | --- | --- | | 2026-09-08 | [Veradigm Inc.](/incidents/veradigm-vendor-api-2026/) | not public | The filing says a third-party vendor had an incident and an unauthorized party obtained credentials from the vendor’s environment to a company application programming interface, then used them to download patient data; the compromised credentials gave access only through that interface. | | 2026-08-06 | [IEH Corporation](/incidents/ieh-microsoft-365-2026/) | Microsoft 365 | The filing says a phishing email carried a link disguised as a Microsoft document-sharing link, and the user entered Microsoft 365 credentials into a fraudulent login page. | | 2026-07-02 | [AdaptHealth Corp.](/incidents/adapthealth-cloud-apps-2026/) | not public | The filing says a threat actor gained unauthorized access to certain cloud-based business applications (the provider is not named) after a social engineering attack that compromised a user session associated with a third-party contractor. | | 2026-07-02 | [AdaptHealth](/incidents/adapthealth-contractor-session-2026/) | not public | The Form 8-K says a threat actor gained unauthorized access to certain cloud-based business applications, and that the incident was the result of a successful social engineering attack that compromised a user session associated with a third-party contractor. | | 2026-06-15 | [iRhythm Holdings, Inc.](/incidents/irhythm-hosted-apps-2026/) | not public | The filing says the affected data was obtained through social engineering from certain third-party-hosted business applications (the provider is not named). | | 2026-05-01 | [A Maryland medical system (called Company A in the indictment release)](/incidents/maryland-medical-system-m365-2016/) | Microsoft 365 | The release says an employee used keylogging, cookie managers and mailbox rules to reach victims’ personal and professional accounts, including Microsoft 365, and exported browser cookies to keep unauthorized access from other devices. | | 2025-11-27 | [Mixpanel](/incidents/mixpanel-smishing-2025/) | Mixpanel | The company stated it detected a smishing campaign on November 8, 2025, and took steps to contain unauthorized access and secure impacted user accounts; it revoked sessions and rotated compromised credentials for impacted accounts. | | 2025-11-06 | [Illuminate Education, Inc.](/incidents/illuminate-education-2021/) | not public (the FTC says a third-party cloud provider, not named in the press release) | The FTC press release says that, according to the complaint, a hacker used the credentials of a former employee, who had left the company three and a half years earlier, to breach databases stored on a third-party cloud provider. It also says a third-party vendor had alerted the company in January 2020 to numerous security vulnerabilities and that student data was stored in plain text until at least January 2022. | | 2025-09-22 | [GitHub (npm registry)](/incidents/github-npm-shai-hulud-2025/) | GitHub | GitHub stated that a self-replicating worm spread through compromised maintainer accounts by injecting malicious post-install scripts into popular packages. | | 2025-08-05 | [Google](/incidents/google-salesforce-instance-2025/) | Salesforce | Google stated that the instance was impacted by UNC6040 activity described in the same post, which describes voice phishing in which employees are tricked into authorizing a malicious connected app or sharing credentials. | | 2025-05-01 | [The Walt Disney Company](/incidents/disney-slack-2024/) | Slack | The release says a malicious file posted online let an individual into an employee’s personal computer, where stored login credentials were obtained and used to reach the employee’s Slack account. | | 2025-04-28 | [Nova Scotia Power](/incidents/nova-scotia-power-opc-2025/) | not public | The letter says an employee visited a compromised website and clicked a link, malware was installed, the threat actor moved across systems using accounts with domain administrator privileges, harvested credentials, took data between 23 and 25 April 2025, then used harvested credentials to destroy backups and deploy ransomware. | | 2025-03-27 | [Advanced Computer Software Group Ltd (health and care subsidiary)](/incidents/advanced-computer-software-2022/) | not public | The ICO states hackers accessed certain systems of the company’s health and care subsidiary via a customer account that did not have multi-factor authentication (MFA). The ICO’s investigation found gaps in the deployment of MFA, a lack of comprehensive vulnerability scanning and inadequate patch management. The ICO does not state how the account was obtained. | | 2025-03-12 | [Salesforce customers (multiple organizations)](/incidents/salesforce-customers-vishing-2025/) | Salesforce | Salesforce stated threat actors impersonated IT support by phone, luring customers' employees and third-party support workers to phishing pages to steal credentials and MFA tokens, or to add a malicious connected app. | | 2025-01-07 | [PowerSchool](/incidents/powerschool-powersource-2024/) | PowerSchool | The company stated, citing its forensic firm’s final report, that the threat actor accessed PowerSource, a community-focused customer support portal, using a single compromised credential. | | 2024-05-30 | [Snowflake customer accounts (multiple organizations)](/incidents/snowflake-customer-accounts-2024/) | Snowflake | Snowflake stated in its annual report that a threat actor accessed a number of customers’ Snowflake accounts because those customers had not met their obligations to implement multi-factor authentication and network access policies, and that it found no breach of its own platform or environment. | | 2024-04-22 | [Payroll2U Pte. Ltd.](/incidents/payroll2u-pdpc-2022/) | AWS | The decision says a single compromised user account was used for Remote Desktop Protocol access to five servers in the company's AWS environment, and that investigators could not determine how the credentials were obtained. It lists no MFA for administrator accounts or accounts with access to personal data. | | 2024-01-19 | [Microsoft](/incidents/microsoft-midnight-blizzard-2024/) | Microsoft 365 | Microsoft stated the threat actor used a password spray attack to compromise a legacy non-production test tenant account, then used that account's permissions to access corporate email accounts. | | 2023-10-20 | [Okta](/incidents/okta-support-system-2023/) | Okta | Okta stated the access used a service account stored in the support system itself, whose username and password had been saved in an employee's personal Google account on an Okta-managed laptop. | | 2023-10-20 | [BeyondTrust](/incidents/beyondtrust-okta-2023/) | Okta | The company stated it detected an attacker trying to access an in-house Okta administrator account using a valid session cookie stolen from the provider’s support system. | | 2023-10-06 | [23andMe](/incidents/23andme-2023/) | not public | The company stated the threat actor accessed accounts through credential stuffing, using usernames and passwords that were also used on other websites previously compromised. | | 2023-08-29 | [Retool](/incidents/retool-2023/) | Okta | The company stated an employee was targeted by an SMS-based phishing attack followed by a call, and that the attacker added their own device to the employee's Okta account and used synced MFA codes to reach the VPN and internal admin systems. | | 2023-07-12 | [JumpCloud](/incidents/jumpcloud-orchestration-2023/) | JumpCloud | The company stated it traced anomalous activity on an internal orchestration system to a spear-phishing campaign on June 22, which gave unauthorized access to a specific area of its infrastructure. It later found unusual activity in the commands framework for a small set of customers, and stated the attack vector was data injection into its commands framework. | | 2023-01-13 | [Mailchimp (Intuit)](/incidents/mailchimp-2023/) | Mailchimp | The company stated an unauthorized actor used a social engineering attack on employees and contractors and used the compromised employee credentials to reach an internal customer support and account administration tool. | | 2023-01-04 | [CircleCI](/incidents/circleci-2023/) | CircleCI | The company stated that malware on an engineer's laptop stole a valid 2FA-backed SSO session, and that the employee held privileges to generate production access tokens. | | 2022-12-31 | [Slack](/incidents/slack-github-tokens-2022/) | GitHub | The company stated that a limited number of employee tokens were stolen and misused to reach its externally hosted GitHub repository, and that a third-party vendor was compromised. | | 2022-12-19 | [RedMart Limited](/incidents/redmart-pdpc-2020/) | AWS and Alibaba Cloud | The decision says a threat actor gained unauthorised access to the GitHub account of a member of the DevOps team, found a high-privilege key to the AWS environment in the repositories, used it to find a second key in a private S3 bucket, created an EC2 instance, and crossed a VPN link to the Alibaba cloud database. The database was neither encrypted nor password protected. | | 2022-11-01 | [Dropbox](/incidents/dropbox-github-phishing-2022/) | GitHub | The company stated that employees received phishing emails impersonating a CI service, and that a fake login page collected GitHub credentials and a one-time code, giving the threat actor access to one GitHub organization. | | 2022-10-24 | [Drizly](/incidents/drizly-ftc-2022/) | AWS | The FTC complaint stated that in early July 2020 a malicious actor reused credentials from an unrelated breach to reach an executive's GitHub account, found AWS and database credentials stored in the company's GitHub repositories, and used them to change the company's AWS security settings and reach the production environment. | | 2022-09-16 | [Uber](/incidents/uber-2022/) | Google Workspace | The company stated that an attacker likely bought a contractor's corporate password after malware on the contractor's personal device exposed it, then repeated two-factor login requests until the contractor accepted one. | | 2022-09-16 | [GitHub users (phishing campaign reported by GitHub)](/incidents/github-circleci-phishing-2022/) | GitHub | GitHub stated that a phishing site imitating the GitHub login page stole credentials and relayed TOTP codes to the threat actor in real time, and that accounts protected by hardware security keys were not vulnerable. | | 2022-08-12 | [Mailchimp (Intuit)](/incidents/mailchimp-crypto-social-engineering-2022/) | Mailchimp | The company stated an unauthorized actor accessed an internal tool used by customer-facing teams for customer support and account administration, after a social engineering attack on Mailchimp employees that obtained employee credentials. | | 2022-08-10 | [Cisco Systems](/incidents/cisco-google-account-vishing-2022/) | Box | The company stated an employee's credentials were compromised after an attacker gained control of a personal Google account to which credentials saved in the browser were synchronised, and the attacker then used voice phishing and repeated multi-factor push requests until one was accepted, which gave access to the VPN. | | 2022-08-09 | [Cloudflare](/incidents/cloudflare-phishing-2022/) | Okta | The company stated that employees received text messages pointing to a look-alike page for its identity provider, which asked for a username, password and a one-time code; individual employees did enter credentials, but physical security keys required for its applications stopped the attacker. | | 2022-08-07 | [Twilio](/incidents/twilio-2022/) | Twilio | The company stated that text messages sent to current and former employees led some of them to a page impersonating the Twilio sign-in page, and that the attackers used the stolen credentials to reach internal systems. | | 2022-07-14 | [Quoine Pte Ltd](/incidents/quoine-pdpc-2020/) | not public | The decision says social engineering of employees of the company's domain registrar led to control of its domain hosting account being handed to an external actor, who redirected the company's mail, reset the password of a DevOps account, and used the reset credentials to reach the company's cloud computing platform, which held API keys to its databases. | | 2022-05-19 | [MGM Resorts International](/incidents/mgm-resorts-opc-2019/) | not public | The OPC finding, relaying MGM's account, says an unauthorized actor logged into a third-party platform used by MGM software developers with an MGM employee's credentials compromised in an earlier unrelated breach, exploited a coding flaw on that platform to bypass multi-factor authentication, found a temporary access token, and used it to reach guest data on an external cloud server for about one hour. | | 2022-04-21 | [GeniusU Pte. Ltd.](/incidents/geniusu-pdpc-2021/) | GitHub | The decision says the likely cause was a compromise of a developer's GitHub password, which let the threat actor into the company's GitHub environment; database login credentials had been stored in the codebase there, so the actor reached and exfiltrated the database. The staging database held real personal data. | | 2022-03-22 | [Microsoft](/incidents/microsoft-dev-0537-2022/) | not public | The company stated that a single account was compromised, granting limited access, and that the tactics used in this intrusion reflect those described in the same post (including stolen credentials and MFA prompt abuse). | | 2021-11-22 | [GoDaddy](/incidents/godaddy-managed-wordpress-2021/) | GoDaddy Managed WordPress hosting | The company stated that an unauthorized third party used a compromised password to access the provisioning system in the legacy code base for Managed WordPress. | | 2021-10-14 | [The National Kidney Foundation](/incidents/national-kidney-foundation-pdpc-2020/) | Microsoft 365 | The decision says an employee received a phishing email linking to a website that asked for account credentials; the hacker is believed to have obtained them, accessed the employee's mailbox, which was hosted on Microsoft Office 365 and reachable by webmail, synchronised it on 17 May 2020, and used it to send phishing emails to 1,039 external contacts. | | 2021-08-30 | [Cetera Entities (Cetera Advisor Networks LLC, Cetera Investment Services LLC, Cetera Financial Specialists LLC, Cetera Advisors LLC, Cetera Investment Advisers LLC)](/incidents/cetera-email-account-takeovers-2017/) | not public | The SEC states the cloud-based email accounts of over 60 personnel were taken over by unauthorized third parties, and that none of the taken over accounts were protected in a manner consistent with the firms’ policies. The press release does not name the email provider or state how the accounts were taken over. | | 2021-08-30 | [Cambridge Investment Research Inc. and Cambridge Investment Research Advisors Inc.](/incidents/cambridge-email-account-takeovers-2018/) | not public | The SEC states the cloud-based email accounts of over 121 representatives were taken over by unauthorized third parties. The SEC order finds the firm discovered the first takeover in January 2018 but did not adopt and implement firm-wide enhanced security measures for those accounts until 2021. The press release does not name the email provider or state how the accounts were taken over. | | 2021-08-30 | [KMS Financial Services Inc.](/incidents/kms-financial-email-account-takeovers-2018/) | not public | The SEC states the cloud-based email accounts of 15 financial advisers or their assistants were taken over by unauthorized third parties. The SEC order finds the firm did not adopt written policies requiring additional firm-wide security measures until May 2020 and did not fully implement them firm-wide until August 2020. The press release does not name the email provider or state how the accounts were taken over. | | 2020-12-14 | [SolarWinds](/incidents/solarwinds-2020/) | Microsoft 365 | The filing says the vulnerability was introduced as a result of a compromise of the Orion software build system, and that the company was made aware of an attack vector used to compromise its Office 365 emails. The Microsoft post says that in actions observed in the Microsoft cloud, attackers gained administrative access either using compromised privileged account credentials (for example stolen passwords) or by forging SAML tokens using compromised token signing certificates. | | 2020-09-10 | [Singapore Medical Association](/incidents/singapore-medical-association-pdpc-2020/) | Microsoft 365 | The decision says the association believed an unauthorised user entered a Microsoft Office 365 email account by a brute force attack but had no logs to confirm, then created a rule forwarding received emails to an external address. The account password had not been changed since November 2013 and there were no measures against repeated login attempts. | | 2020-07-18 | [Twitter](/incidents/twitter-2020/) | Twitter | The company says a phone spear phishing attack targeted a small number of employees; the attackers used the employees' credentials to access internal systems, including getting through two-factor protections, then used the credentials of employees with access to account support tools. | ## The kind of control that addresses it require phishing-resistant multifactor authentication for every person with access to the data platform, and limit what any single sign-in can reach ## Related pages - [Over-permissioned identities: cloud breaches by access pat](/access-paths/over-permissioned-identity/) - [Public storage, databases and services: cloud breaches by ](/access-paths/public-resource/) - [Cross-tenant access: cloud breaches by access path](/access-paths/cross-tenant/) ## Sources - Microsoft Learn: how multifactor authentication works: https://learn.microsoft.com/en-us/entra/identity/authentication/concept-mfa-howitworks (read 16 Nov 2026) - AWS IAM User Guide: multi-factor authentication in IAM: https://docs.aws.amazon.com/IAM/latest/UserGuide/id_credentials_mfa.html (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/access-paths/stolen-user-credentials/ --- # Over-permissioned identities: 13 incidents in the ledger Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. An over-permissioned identity is a user, role, service account or application that held more access than its job needed, so that the extra access, and not the way the identity was compromised, set how much data or infrastructure an incident reached and how long the intruder could keep moving. ## How it works Providers state the intent plainly. Microsoft's documentation says least privilege means granting administrators exactly the permission they need to do their job, over a specific scope and for a specific time. AWS documentation describes a permissions boundary as a policy that sets the maximum permissions an identity-based policy can grant to a user or role. These are limits an owner has to apply, because identities are created with whatever access the person creating them grants. In an incident, the identity that is compromised is often not the problem by itself. What the sources record is how far its permissions reached: a role that could read every bucket, or an account with administrative rights where a narrow one would do. AWS documentation also tells owners to find unused credentials and remove them, because access that is never used still counts when the identity is stolen. ## The incidents | Disclosed | Organization | Cloud | How access was gained | | --- | --- | --- | --- | | 2025-08-26 | [Nx (Nrwl)](/incidents/nx-github-actions-2025/) | GitHub | The post-mortem says attackers exploited a GitHub Actions injection vulnerability to steal the npm publishing token; the repository's workflow permissions were read and write. | | 2025-07-23 | [Amazon Web Services](/incidents/aws-amazon-q-extension-2025/) | AWS | The bulletin says the extension had an inappropriately scoped GitHub token in its CodeBuild configuration, and with that token the threat actor committed malicious code. | | 2025-04-28 | [Nova Scotia Power](/incidents/nova-scotia-power-opc-2025/) | not public | The letter says an employee visited a compromised website and clicked a link, malware was installed, the threat actor moved across systems using accounts with domain administrator privileges, harvested credentials, took data between 23 and 25 April 2025, then used harvested credentials to destroy backups and deploy ransomware. | | 2024-08-30 | [Verkada Inc.](/incidents/verkada-2021/) | Verkada (cloud-based security camera provider) | The FTC press release says the complaint alleges the company failed to provide appropriate security measures, including failing to require unique and complex passwords, adequately encrypt customer data and implement secure network controls. The Justice Department release says the complaint alleges failures in appropriate access management and data protection controls. Neither source states how the hacker obtained the access. | | 2024-05-01 | [Dropbox](/incidents/dropbox-sign-2024/) | not public | The company stated a third party gained access to an automated system configuration tool, compromised a back-end service account with privileges across the production environment, and reached the customer database; it answered how access was gained with a compromised access token. | | 2024-01-19 | [Microsoft](/incidents/microsoft-midnight-blizzard-2024/) | Microsoft 365 | Microsoft stated the threat actor used a password spray attack to compromise a legacy non-production test tenant account, then used that account's permissions to access corporate email accounts. | | 2023-09-18 | [Microsoft](/incidents/microsoft-ai-sas-token-2023/) | Microsoft Azure | Microsoft stated an employee shared a URL for a blob store in a public GitHub repository, and the URL included an overly-permissive Shared Access Signature token for an internal storage account. | | 2023-08-16 | [E-Commerce Enablers Pte. Ltd.](/incidents/ecommerce-enablers-pdpc-2020/) | AWS | The decision says an AWS access key with full administrative privileges was committed to a private GitHub repository on 4 June 2019, removed on 6 June but still viewable in the commit history, and not fully disabled when a replacement key was created. On 9 September 2020 a threat actor used it to reach the AWS environment, changed security settings to allow remote internet access to database instances and exfiltrated data. The company conjectured the key was found in the commit history. | | 2023-01-04 | [CircleCI](/incidents/circleci-2023/) | CircleCI | The company stated that malware on an engineer's laptop stole a valid 2FA-backed SSO session, and that the employee held privileges to generate production access tokens. | | 2022-09-16 | [Uber](/incidents/uber-2022/) | Google Workspace | The company stated that an attacker likely bought a contractor's corporate password after malware on the contractor's personal device exposed it, then repeated two-factor login requests until the contractor accepted one. | | 2022-08-12 | [Mailchimp (Intuit)](/incidents/mailchimp-crypto-social-engineering-2022/) | Mailchimp | The company stated an unauthorized actor accessed an internal tool used by customer-facing teams for customer support and account administration, after a social engineering attack on Mailchimp employees that obtained employee credentials. | | 2021-12-01 | [A publicly traded wireless communications manufacturer (called Company-1 in the U.S. Attorney release)](/incidents/ubiquiti-aws-github-2020/) | AWS | The release says an employee who was a senior developer held credentials for the company’s AWS and GitHub servers and repeatedly misused his administrative access to download the data; the company’s own account of an outside intruder was, the release says, false. | | 2020-07-16 | [Blackbaud Inc.](/incidents/blackbaud-2020/) | Blackbaud (software provider; the New York Attorney General calls it a cloud company) | The FTC press release says that, according to the complaint, a hacker in early 2020 accessed a customer’s Blackbaud-hosted database and, once logged in, moved across multiple Blackbaud-hosted environments by using existing vulnerabilities and local administrator accounts and by creating new administrator accounts. The FTC says the company failed to segment data, adequately implement multifactor authentication and monitor for breach attempts. The company paid a ransom of 24 Bitcoin, according to the FTC. | ## The kind of control that addresses it scope each identity to the actions and resources its task needs, cap delegated permissions with boundaries, and remove access that is unused ## Related pages - [Public storage, databases and services: cloud breaches by ](/access-paths/public-resource/) - [Cross-tenant access: cloud breaches by access path](/access-paths/cross-tenant/) - [Third-party access: cloud breaches by access path](/access-paths/third-party-access/) ## Sources - Microsoft Learn: best practices for Microsoft Entra roles: https://learn.microsoft.com/en-us/entra/identity/role-based-access-control/best-practices (read 16 Nov 2026) - AWS IAM User Guide: permissions boundaries: https://docs.aws.amazon.com/IAM/latest/UserGuide/access_policies_boundaries.html (read 16 Nov 2026) - AWS IAM User Guide: find unused credentials: https://docs.aws.amazon.com/IAM/latest/UserGuide/id_credentials_finding-unused.html (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/access-paths/over-permissioned-identity/ --- # Public storage, databases and services: 15 incidents in the ledger Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. A public resource is a storage container, database or service that was left reachable to anyone who could find its address, with no sign-in required, because a setting, a policy or a firewall rule allowed anonymous access, whether by choice, by mistake or through a later change. ## How it works Provider defaults are closed. Amazon's documentation says new buckets, access points and objects do not allow public access by default, and that users can change policies or permissions to allow it. Microsoft's documentation says anonymous access to blob data is prohibited by default and that a container set for anonymous access can be read by any client. Public access is therefore a setting someone changed, or a policy attached later. Because the resource asks for no credentials, the only barrier is the address, and addresses are guessed or scanned. Both providers document an account-wide switch that overrides individual settings: Amazon's documentation describes a block that applies at the organization, account and bucket levels, and Microsoft describes disallowing anonymous access for a whole storage account. Google Cloud documents a public access prevention setting for the same purpose. ## The incidents | Disclosed | Organization | Cloud | How access was gained | | --- | --- | --- | --- | | 2023-11-16 | [Global Tel*Link Corp.](/incidents/global-tel-link-ftc-2020/) | not public | The FTC press release says that, after the company and a third-party vendor copied the data into the cloud for testing, changes made by the vendor to the security settings left the data accessible via the internet without safeguards, until a security researcher alerted the company. | | 2023-06-16 | [1Health.io Inc. (formerly Vitagene, Inc.)](/incidents/vitagene-1health-ftc-2019/) | AWS | The FTC press release says the company stored the data in publicly accessible buckets on AWS cloud storage, did not encrypt it, restrict access to it, log or monitor access to it, or inventory it, and that it was warned at least three times over a two-year period before investigating after a security researcher contacted it in June 2019. | | 2023-03-10 | [Eatigo International Pte. Ltd.](/incidents/eatigo-pdpc-2020/) | not public | The decision says the legacy database sat on the infrastructure of a cloud service provider in Singapore, was left out of the company's private network setup, was reachable from the internet to anyone with the credentials, and had no password rotation rule. Staff no longer knew it existed, and no asset inventory or access logs were kept, so the company could not establish how or when access was gained. | | 2022-10-19 | [Microsoft (misconfigured storage endpoint)](/incidents/microsoft-misconfigured-endpoint-2022/) | not public | The post-mortem says an unintentional misconfiguration on an endpoint, not a security vulnerability, allowed unauthenticated access; the endpoint is now accessible only with authentication. | | 2022-07-14 | [Crawfort Pte. Ltd.](/incidents/crawfort-pdpc-2020/) | AWS | The decision says the company opened an S3 server port in its AWS environment for one week during a data migration, and that the S3 bucket holding the data was publicly accessible due to a misconfiguration. It states the threat actor gained access to the public bucket in that period. | | 2022-06-30 | [Wegmans](/incidents/wegmans-azure-containers-2021/) | Microsoft Azure | A security researcher told the company in April 2021 that a cloud storage container hosted on Microsoft Azure was open to public access; the release says a second container had been public since November 2018. | | 2022-06-30 | [Wegmans Food Markets, Inc.](/incidents/wegmans-azure-container-2018/) | Microsoft Azure | The New York Attorney General states that in April 2021 a security researcher informed the company that a cloud storage container hosted on Microsoft Azure was left unsecured and open to public access. The Attorney General’s office found the company failed to configure the container to limit access, to inventory its cloud assets, to secure all user passwords and to regularly conduct security testing of its cloud assets. | | 2021-10-14 | [Stylez Pte Ltd](/incidents/stylez-icompare-pdpc-2019/) | not public | The decision says the testing database was hosted on a cloud server leased from a cloud storage service provider, stored in a publicly accessible directory without access controls, so it was reachable from the internet and was crawled and indexed by search engines. | | 2020-12-16 | [SkyMed International, Inc.](/incidents/skymed-ftc-2020/) | not public | The FTC press release says the database was unsecured and exposed by a security researcher, and that it could be located and accessed by anyone on the Internet. The press release says the company deleted the database after confirming that the data was online and publicly accessible. | | 2020-12-15 | [Ascension Data & Analytics](/incidents/ascension-data-2019/) | not public | The FTC complaint says a vendor hired for text recognition scanning stored the documents on a cloud-based server and a separate cloud-based storage location, misconfigured both, and left the information exposed to anyone on the internet; no password was required. The complaint says the respondent did not assess the vendor's security measures. | | 2020-10-16 | [Chan Brothers Travel Pte Ltd](/incidents/chan-brothers-travel-pdpc-2019/) | Salesforce | The decision says the company hosted its site on Salesforce Community Cloud with the access configuration set to Public. A March 2018 release added an automated search engine optimisation feature, and the forms inherited the public setting, so the data was indexed. The company said it received no notice of the release; the decision found the release notes were published. | | 2020-10-16 | [MRI Diagnostics Pte Ltd and Clarity Radiology Pte Ltd](/incidents/mri-diagnostics-clarity-radiology-pdpc-2018/) | not public | The decision says the clinic subscribed to an internet-based teleradiology system provided by Clarity Radiology, a cloud-based software service, which engaged an overseas vendor to maintain it. The vendor removed access restrictions on a network folder to patch the system and did not restore them, so the spreadsheets were indexed by search engines. | | 2020-07-22 | [Twilio](/incidents/twilio-2020/) | AWS | The company's incident report says a misconfiguration in the S3 bucket hosting the library let a bad actor inject code, and that the access policy for the path storing the SDK allowed anybody on the Internet to read and write to that path. | | 2020-01-22 | [Microsoft](/incidents/microsoft-support-database-2019/) | Microsoft Azure | The post says a change made to the database's network security group on December 5, 2019 contained misconfigured security rules that enabled exposure of the data. It says the investigation found no malicious use, and that the issue was specific to an internal database and does not represent an exposure of the company's commercial cloud services. | | 2019-06-28 | [Online Buddies (Jack’d dating app)](/incidents/online-buddies-jackd-s3-2018/) | AWS | The New York Attorney General’s investigation confirmed the company failed to secure user photos stored in an AWS S3 storage service, and left app interfaces to backend data unsecured. | ## The kind of control that addresses it enforce an account-wide or organization-wide block on public access to storage, and require sign-in for every database and service ## Related pages - [Cross-tenant access: cloud breaches by access path](/access-paths/cross-tenant/) - [Third-party access: cloud breaches by access path](/access-paths/third-party-access/) - [Exposed network services and request paths: cloud breaches](/access-paths/exposed-network-service/) ## Sources - Amazon S3 User Guide: blocking public access: https://docs.aws.amazon.com/AmazonS3/latest/userguide/access-control-block-public-access.html (read 16 Nov 2026) - Microsoft Learn: remediate anonymous read access to blob data: https://learn.microsoft.com/en-us/azure/storage/blobs/anonymous-read-access-prevent (read 16 Nov 2026) - Google Cloud: public access prevention: https://cloud.google.com/storage/docs/public-access-prevention (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/access-paths/public-resource/ --- # Cross-tenant access: 3 incidents in the ledger Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Cross-tenant access is a flaw in a provider's service, or a setting in a customer's application or account, that let one customer's tenant reach resources belonging to another customer, so the boundary the provider draws around each customer did not hold for that request or token. ## How it works A tenant is the boundary a provider draws around one customer. Microsoft's documentation says its identity service organizes users and apps into tenants, and that a multitenant app can be used by accounts from any directory, so the app has to decide whose sign-ins it accepts. Providers run shared infrastructure behind these boundaries, so a flaw in the provider's own service can cross them without any mistake by the customer. The ledger records this path only where the provider or the customer's own statement describes the crossing. Typical cases are an application that accepts a valid token from the wrong tenant, a provider-side service that handled another customer's credentials, or a setting that allowed data to be copied between tenants. Fixes come from the provider when the flaw is theirs and from the app owner when the check is theirs. ## The incidents | Disclosed | Organization | Cloud | How access was gained | | --- | --- | --- | --- | | 2023-07-12 | [JumpCloud](/incidents/jumpcloud-orchestration-2023/) | JumpCloud | The company stated it traced anomalous activity on an internal orchestration system to a spear-phishing campaign on June 22, which gave unauthorized access to a specific area of its infrastructure. It later found unusual activity in the commands framework for a small set of customers, and stated the attack vector was data injection into its commands framework. | | 2023-07-11 | [Microsoft](/incidents/microsoft-storm-0558-2023/) | Microsoft 365 | Microsoft stated the actor used an acquired Microsoft account consumer signing key to forge tokens, and exploited a token validation issue so that enterprise mail accepted them. The post-mortem says the most probable mechanism for the key acquisition was a compromised engineer corporate account with access to a debugging environment holding a crash dump that contained the key. | | 2021-08-27 | [Microsoft (Azure Cosmos DB)](/incidents/azure-cosmosdb-chaosdb-2021/) | Microsoft Azure | The post-mortem says a vulnerability in the Azure Cosmos DB Jupyter Notebook feature could allow a user to gain access to another customer's resources by using the account's primary read-write key. | ## The kind of control that addresses it validate the tenant and issuer on every token an application accepts, and turn off cross-tenant features that are not in use ## Related pages - [Third-party access: cloud breaches by access path](/access-paths/third-party-access/) - [Exposed network services and request paths: cloud breaches](/access-paths/exposed-network-service/) - [Retained access after departure or change: cloud breaches ](/access-paths/retained-access/) ## Sources - Microsoft Learn: single and multitenant apps in Microsoft Entra ID: https://learn.microsoft.com/en-us/entra/identity-platform/single-and-multi-tenant-apps (read 16 Nov 2026) - AWS IAM User Guide: the confused deputy problem: https://docs.aws.amazon.com/IAM/latest/UserGuide/confused-deputy.html (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/access-paths/cross-tenant/ --- # Third-party access: 25 incidents in the ledger Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Third-party access is access to a customer's cloud or software-as-a-service data that came through a vendor, supplier, integration or contractor the customer had trusted, so the incident began in the outside party's systems or credentials and reached the customer through a grant already in place. ## How it works Customers grant outside parties access on purpose: a monitoring company, a data integration, an analytics tool. AWS documentation describes the pattern with a role that trusts another account, and says the external ID in the trust policy is there to prevent the confused deputy problem, in which an entity without permission coerces a more-privileged entity to act for it. The third party often serves many customers from one account, so the trust is shared. When the vendor is compromised, its access to each customer is the path. The sources in the ledger describe tokens held by an integration, a contractor account or a connected app. The customer's exposure is whatever it granted, which is why the documentation tells owners to scope what a third party can reach and to condition the grant on an identifier that is specific to the customer. ## The incidents | Disclosed | Organization | Cloud | How access was gained | | --- | --- | --- | --- | | 2026-09-08 | [Veradigm Inc.](/incidents/veradigm-vendor-api-2026/) | not public | The filing says a third-party vendor had an incident and an unauthorized party obtained credentials from the vendor’s environment to a company application programming interface, then used them to download patient data; the compromised credentials gave access only through that interface. | | 2026-07-02 | [AdaptHealth Corp.](/incidents/adapthealth-cloud-apps-2026/) | not public | The filing says a threat actor gained unauthorized access to certain cloud-based business applications (the provider is not named) after a social engineering attack that compromised a user session associated with a third-party contractor. | | 2026-07-02 | [AdaptHealth](/incidents/adapthealth-contractor-session-2026/) | not public | The Form 8-K says a threat actor gained unauthorized access to certain cloud-based business applications, and that the incident was the result of a successful social engineering attack that compromised a user session associated with a third-party contractor. | | 2026-06-23 | [8x8, Inc.](/incidents/8x8-salesforce-klue-2026/) | Salesforce | The filing says an unauthorized third party threat actor exploited the Klue third-party application programming integration connected to the company's Salesforce customer relationship management system. | | 2026-05-08 | [SailPoint, Inc.](/incidents/sailpoint-github-2026/) | GitHub | The filing says the company detected unauthorized access to a subset of its GitHub repositories and that the root cause was a vulnerability in a third-party application, which it does not name. | | 2025-09-02 | [Cloudflare](/incidents/cloudflare-salesforce-drift-2025/) | Salesforce | The company stated the threat actor obtained OAuth credentials associated with the Salesloft Drift chat agent's Salesforce integration, and used a stolen credential used by the Salesloft integration to log in to its Salesforce tenant. | | 2025-08-20 | [Salesloft (Drift)](/incidents/salesloft-drift-2025/) | AWS | The investigation summary says the threat actor accessed the Salesloft GitHub account (March to June 2025), then accessed Drift's AWS environment and obtained OAuth tokens for customers' integrations, and used them to access data via Drift integrations. | | 2025-08-05 | [Google](/incidents/google-salesforce-instance-2025/) | Salesforce | Google stated that the instance was impacted by UNC6040 activity described in the same post, which describes voice phishing in which employees are tricked into authorizing a malicious connected app or sharing credentials. | | 2025-03-12 | [Salesforce customers (multiple organizations)](/incidents/salesforce-customers-vishing-2025/) | Salesforce | Salesforce stated threat actors impersonated IT support by phone, luring customers' employees and third-party support workers to phishing pages to steal credentials and MFA tokens, or to add a malicious connected app. | | 2024-10-22 | [Avaya Holdings Corp.](/incidents/avaya-cloud-file-sharing-2020/) | not public | The SEC press release says the threat actor likely behind the SolarWinds Orion hack accessed the company’s systems without authorization. It describes the access as part of an investigation of public companies potentially impacted by the compromise of SolarWinds’ Orion software. The SEC charged the company over the accuracy of its public disclosure of the incident. | | 2024-02-01 | [Cloudflare](/incidents/cloudflare-thanksgiving-2023/) | Okta | The company stated the actor used one access token and three service account credentials taken in the October 2023 Okta compromise that the company had failed to rotate. | | 2023-11-16 | [Global Tel*Link Corp.](/incidents/global-tel-link-ftc-2020/) | not public | The FTC press release says that, after the company and a third-party vendor copied the data into the cloud for testing, changes made by the vendor to the security settings left the data accessible via the internet without safeguards, until a security researcher alerted the company. | | 2023-10-23 | [1Password](/incidents/1password-okta-2023/) | Okta | The company stated it detected suspicious activity on its Okta instance used to manage employee-facing apps on September 29, and later confirmed it was a result of the provider’s Support System breach. | | 2023-10-20 | [BeyondTrust](/incidents/beyondtrust-okta-2023/) | Okta | The company stated it detected an attacker trying to access an in-house Okta administrator account using a valid session cookie stolen from the provider’s support system. | | 2022-12-31 | [Slack](/incidents/slack-github-tokens-2022/) | GitHub | The company stated that a limited number of employee tokens were stolen and misused to reach its externally hosted GitHub repository, and that a third-party vendor was compromised. | | 2022-07-14 | [Quoine Pte Ltd](/incidents/quoine-pdpc-2020/) | not public | The decision says social engineering of employees of the company's domain registrar led to control of its domain hosting account being handed to an external actor, who redirected the company's mail, reset the password of a DevOps account, and used the reset credentials to reach the company's cloud computing platform, which held API keys to its databases. | | 2022-05-19 | [MGM Resorts International](/incidents/mgm-resorts-opc-2019/) | not public | The OPC finding, relaying MGM's account, says an unauthorized actor logged into a third-party platform used by MGM software developers with an MGM employee's credentials compromised in an earlier unrelated breach, exploited a coding flaw on that platform to bypass multi-factor authentication, found a temporary access token, and used it to reach guest data on an external cloud server for about one hour. | | 2022-04-15 | [Heroku (Salesforce) and Travis CI OAuth tokens](/incidents/heroku-travisci-oauth-2022/) | GitHub | Heroku stated that initial access to its environment was obtained by using a compromised token for a Heroku machine account, after which stored customer GitHub integration OAuth tokens were downloaded; GitHub stated the tokens were then abused against GitHub.com. | | 2022-03-22 | [Okta (contractor Sitel)](/incidents/okta-sitel-2022/) | Okta | The post-mortem says an attacker obtained remote access, using RDP, to a computer owned and managed by Okta's sub-processor Sitel and used by a customer support engineer; the attacker did not take over an Okta account. | | 2021-04-15 | [Codecov](/incidents/codecov-2021/) | Codecov | The company stated that an error in its Docker image creation process allowed the actor to extract the credential required to modify the Bash Uploader script, which was then altered to send CI environment variables to a third-party server. | | 2021-03-16 | [Mimecast](/incidents/mimecast-solarwinds-certificate-2021/) | Microsoft 365 | The company stated the threat actor used the SolarWinds supply-chain compromise to gain access to part of its production grid environment, and from there accessed Mimecast-issued certificates and related connection information; the certificate was then used against a small number of customer Microsoft 365 tenants. | | 2020-12-15 | [Ascension Data & Analytics](/incidents/ascension-data-2019/) | not public | The FTC complaint says a vendor hired for text recognition scanning stored the documents on a cloud-based server and a separate cloud-based storage location, misconfigured both, and left the information exposed to anyone on the internet; no password was required. The complaint says the respondent did not assess the vendor's security measures. | | 2020-12-14 | [SolarWinds](/incidents/solarwinds-2020/) | Microsoft 365 | The filing says the vulnerability was introduced as a result of a compromise of the Orion software build system, and that the company was made aware of an attack vector used to compromise its Office 365 emails. The Microsoft post says that in actions observed in the Microsoft cloud, attackers gained administrative access either using compromised privileged account credentials (for example stolen passwords) or by forging SAML tokens using compromised token signing certificates. | | 2020-10-16 | [MRI Diagnostics Pte Ltd and Clarity Radiology Pte Ltd](/incidents/mri-diagnostics-clarity-radiology-pdpc-2018/) | not public | The decision says the clinic subscribed to an internet-based teleradiology system provided by Clarity Radiology, a cloud-based software service, which engaged an overseas vendor to maintain it. The vendor removed access restrictions on a network folder to patch the system and did not restore them, so the spreadsheets were indexed by search engines. | | 2020-07-11 | [LiveAuctioneers](/incidents/liveauctioneers-2020/) | AWS | The company's letter to the Washington Attorney General says a hacker breached one of its third-party software solutions and obtained a source control key, which gave access to its source repository, its cloud-based storage environment and its production database, then used internal users' credentials to download the user database. | ## The kind of control that addresses it grant each vendor a scoped role bound to a customer-specific condition, review integrations on a schedule, and revoke tokens the vendor no longer needs ## Related pages - [Exposed network services and request paths: cloud breaches](/access-paths/exposed-network-service/) - [Retained access after departure or change: cloud breaches ](/access-paths/retained-access/) - [Leaked access keys and secrets: cloud breaches by access p](/access-paths/leaked-access-key/) ## Sources - AWS IAM User Guide: access to accounts owned by third parties: https://docs.aws.amazon.com/IAM/latest/UserGuide/id_roles_common-scenarios_third-party.html (read 16 Nov 2026) - AWS IAM User Guide: the confused deputy problem: https://docs.aws.amazon.com/IAM/latest/UserGuide/confused-deputy.html (read 16 Nov 2026) - Microsoft Learn: manage application permissions: https://learn.microsoft.com/en-us/entra/identity/enterprise-apps/manage-application-permissions (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/access-paths/third-party-access/ --- # Exposed network services and request paths: 8 incidents in the ledger Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. An exposed network service is a reachable service or request path, such as an open management port, an unpatched internet-facing application or a request forged to reach a metadata endpoint, that gave a caller a way into a workload and into the credentials or data attached to it. ## How it works Every cloud virtual machine can reach a local metadata service that hands out temporary credentials. AWS documentation says the service has two versions, that by default an instance can use either, and that an instance can be set to accept only the session-based version. A request path in an application that fetches a URL on a caller's behalf can be steered at that local address, and the credentials in the reply belong to the instance's role. Open ports follow the same logic. Network rules at each provider allow or deny traffic by source, port and protocol, and Microsoft's documentation describes a default deny rule for inbound traffic that explicit rules override. A rule that opens a management port to the whole internet makes the service behind it the first step of an incident. Sources in the ledger name the service or request path when the company or provider does. ## The incidents | Disclosed | Organization | Cloud | How access was gained | | --- | --- | --- | --- | | 2026-01-08 | [People Central Pte. Ltd.](/incidents/people-central-pdpc-2024/) | AWS | The decision says the company, a cloud-based HR software provider, had Remote Desktop Protocol access open to the internet without two-factor authentication on its AWS cloud servers, settings allowed inbound traffic from all IP addresses in the cloud environment, and SQL injection vulnerabilities were present with no web application firewall. | | 2025-03-20 | [People Central Pte. Ltd. (Singapore, cloud-based HR software provider)](/incidents/people-central-aws-rdp-2024/) | Amazon Web Services | The decision lists contributing lapses: SQL injection vulnerabilities in the web application with no web application firewall, RDP access open to the internet without two-factor authentication, settings that allowed inbound traffic from all IP addresses in the cloud environment, and vulnerability scanning only every two years. | | 2024-12-08 | [BeyondTrust](/incidents/beyondtrust-remote-support-saas-2024/) | Amazon Web Services | The company stated a zero-day vulnerability of a third-party application was used to gain access to an online asset in a BeyondTrust AWS account, which allowed the threat actor to obtain an infrastructure API key that was then used against a separate AWS account that operated Remote Support infrastructure. | | 2024-08-30 | [Verkada Inc.](/incidents/verkada-2021/) | Verkada (cloud-based security camera provider) | The FTC press release says the complaint alleges the company failed to provide appropriate security measures, including failing to require unique and complex passwords, adequately encrypt customer data and implement secure network controls. The Justice Department release says the complaint alleges failures in appropriate access management and data protection controls. Neither source states how the hacker obtained the access. | | 2024-04-22 | [Payroll2U Pte. Ltd.](/incidents/payroll2u-pdpc-2022/) | AWS | The decision says a single compromised user account was used for Remote Desktop Protocol access to five servers in the company's AWS environment, and that investigators could not determine how the credentials were obtained. It lists no MFA for administrator accounts or accounts with access to personal data. | | 2022-07-14 | [Crawfort Pte. Ltd.](/incidents/crawfort-pdpc-2020/) | AWS | The decision says the company opened an S3 server port in its AWS environment for one week during a data migration, and that the S3 bucket holding the data was publicly accessible due to a misconfiguration. It states the threat actor gained access to the public bucket in that period. | | 2019-08-27 | [Imperva](/incidents/imperva-2019/) | AWS | The company's follow-up post says an unauthorized use of an administrative API key in one of its production AWS accounts led to exposure of a database snapshot. It lists the steps as a snapshot created for testing, an internal compute instance accessible from the outside world that contained an AWS API key, compromise of that instance and theft of the key, and use of the key to access the snapshot. | | 2019-07-29 | [Capital One](/incidents/capital-one-2019/) | AWS | The company stated that a highly sophisticated individual was able to exploit a specific configuration vulnerability in its infrastructure and that the issue was not specific to the cloud. The OCC consent order finds the bank failed to establish appropriate risk management for its cloud operating environment, including network security controls. | ## The kind of control that addresses it require session-based metadata access on every instance, close management ports to the internet, and limit the role attached to each workload ## Related pages - [Retained access after departure or change: cloud breaches ](/access-paths/retained-access/) - [Leaked access keys and secrets: cloud breaches by access p](/access-paths/leaked-access-key/) - [Stolen user credentials and session tokens: cloud breaches](/access-paths/stolen-user-credentials/) ## Sources - Amazon EC2 User Guide: the Instance Metadata Service: https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/configuring-instance-metadata-service.html (read 16 Nov 2026) - Microsoft Learn: network security groups overview: https://learn.microsoft.com/en-us/azure/virtual-network/network-security-groups-overview (read 16 Nov 2026) - Google Cloud: VPC firewall rules: https://cloud.google.com/vpc/docs/firewalls (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/access-paths/exposed-network-service/ --- # Retained access after departure or change: 7 incidents in the ledger Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Retained access is access that a former employee, contractor or application kept after the reason for it ended, such as a password, key or connected app that nobody removed, and which was later used or left available to be used against the organization's cloud accounts. ## How it works Access is granted when a person joins a project and is removed only if someone acts. AWS documentation says that when users leave an organization or no longer need access, the credentials they used should be found and made non-operational, and at a minimum the password changed or the access keys deactivated. It describes finding credentials by last-used dates, since the definition of unused is a period without use. The same holds for applications and tokens. Microsoft's guidance on Midnight Blizzard, read 16 Nov 2026, says malicious OAuth applications let a threat actor keep access even after losing the account that created them. In both cases the preventive control is an offboarding step that covers every credential and grant a person or app created, and a periodic review that finds the ones nobody uses. ## The incidents | Disclosed | Organization | Cloud | How access was gained | | --- | --- | --- | --- | | 2026-05-01 | [A Maryland medical system (called Company A in the indictment release)](/incidents/maryland-medical-system-m365-2016/) | Microsoft 365 | The release says an employee used keylogging, cookie managers and mailbox rules to reach victims’ personal and professional accounts, including Microsoft 365, and exported browser cookies to keep unauthorized access from other devices. | | 2025-11-06 | [Illuminate Education, Inc.](/incidents/illuminate-education-2021/) | not public (the FTC says a third-party cloud provider, not named in the press release) | The FTC press release says that, according to the complaint, a hacker used the credentials of a former employee, who had left the company three and a half years earlier, to breach databases stored on a third-party cloud provider. It also says a third-party vendor had alerted the company in January 2020 to numerous security vulnerabilities and that student data was stored in plain text until at least January 2022. | | 2023-12-11 | [A bank headquartered in San Francisco (not named in the U.S. Attorney release)](/incidents/sf-bank-cloud-engineer-retained-laptop-2020/) | not public | The release says a cloud engineer fired on March 11, 2020 used a company-issued laptop he had not returned to access the bank's computer network without authorization. | | 2022-10-31 | [Chegg](/incidents/chegg-ftc-2022/) | not public | The FTC stated that a former contractor used login information the company shared with employees and outside contractors to access one of Chegg's third-party cloud databases. | | 2022-02-23 | [Alfi, Inc.](/incidents/alfi-former-executives-cloud-lockout-2022/) | Amazon Web Services | The filing says that after the executives were placed on leave and told not to access the company's computer systems or platforms, one of them took control of the Google Workspace and Amazon Web Services accounts, and two others accessed and downloaded files from Google Drive. | | 2021-09-21 | [Sendtech Pte. Ltd.](/incidents/sendtech-pdpc-2021/) | AWS | The decision says the cause was a compromised AWS access key created in 2015 and not rotated since. With it the attacker gained admin privileges, created another admin account and queried the buckets storing personal data. The company suspected the key could have been compromised through former or current employees but the decision does not confirm how. | | 2020-08-26 | [Cisco Systems](/incidents/cisco-webex-aws-2018/) | AWS | The Department of Justice says that under the plea agreement an individual who had worked for Cisco and resigned in approximately April 2018 admitted intentionally accessing Cisco's cloud infrastructure hosted by Amazon Web Services without permission on September 24, 2018, and deployed code from a Google Cloud Project account that deleted the virtual machines. | ## The kind of control that addresses it remove every credential and grant tied to a departing person or retired app on the day it ends, and review unused access on a fixed schedule ## Related pages - [Leaked access keys and secrets: cloud breaches by access p](/access-paths/leaked-access-key/) - [Stolen user credentials and session tokens: cloud breaches](/access-paths/stolen-user-credentials/) - [Over-permissioned identities: cloud breaches by access pat](/access-paths/over-permissioned-identity/) ## Sources - AWS IAM User Guide: find unused credentials: https://docs.aws.amazon.com/IAM/latest/UserGuide/id_credentials_finding-unused.html (read 16 Nov 2026) - Microsoft Security Blog: Midnight Blizzard guidance for responders: https://www.microsoft.com/en-us/security/blog/2024/01/25/midnight-blizzard-guidance-for-responders-on-nation-state-attack/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/access-paths/retained-access/ --- # Cloud breaches by cloud and provider Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. The 111 incidents in the ledger involve 7 clouds or providers as their primary sources name them; "not public" marks an incident whose sources name none. Each page lists the incidents with the access path used and links each entry to its sources, counted on 16 Nov 2026. A provider with three or more incidents in the ledger has its own page; the others share one page with the incidents whose sources name no provider. The provider is the one the company statement, filing, court document or post-mortem names, which can be the SaaS platform (for example Snowflake or Salesforce) rather than the cloud it runs on. | Cloud or provider | Incidents | | --- | --- | | [AWS](/clouds/aws/) | 19 | | [Amazon Web Services](/clouds/amazon-web-services/) | 3 | | [GitHub](/clouds/github/) | 9 | | [Microsoft 365](/clouds/microsoft-365/) | 9 | | [Microsoft Azure](/clouds/microsoft-azure/) | 5 | | [Okta](/clouds/okta/) | 7 | | [Salesforce](/clouds/salesforce/) | 5 | | [Other providers and not public](/clouds/other/) | 54 | ## Related pages - [Cloud breaches by year, 2019 to 2026](/years/) - [Disclosed cloud attack techniques, with sources](/techniques/) - [Guides](/guides/) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/clouds/ --- # Other providers, and incidents whose provider is not public: 54 Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. 54 incidents in the ledger involve a cloud or SaaS provider with fewer than three entries, or a provider that no primary source names (shown as not public). Each row gives the provider as the sources state it, the access path the ledger reads and a link to the entry with its primary sources, counted on 16 Nov 2026. | Disclosed | Organization | Cloud or provider | Access path | How access was gained | | --- | --- | --- | --- | --- | | 2026-09-08 | [Veradigm Inc.](/incidents/veradigm-vendor-api-2026/) | not public | Third-party access | The filing says a third-party vendor had an incident and an unauthorized party obtained credentials from the vendor’s environment to a company application programming interface, then used them to download patient data; the compromised credentials gave access only through that interface. | | 2026-07-31 | [Amgen](/incidents/amgen-cloud-environments-2026/) | not public | Not public | The Form 8-K says the company identified unauthorized activity involving data stored in cloud environments hosted by third-party cloud service providers. How access was gained is not public. | | 2026-07-02 | [AdaptHealth Corp.](/incidents/adapthealth-cloud-apps-2026/) | not public | Third-party access | The filing says a threat actor gained unauthorized access to certain cloud-based business applications (the provider is not named) after a social engineering attack that compromised a user session associated with a third-party contractor. | | 2026-07-02 | [AdaptHealth](/incidents/adapthealth-contractor-session-2026/) | not public | Third-party access | The Form 8-K says a threat actor gained unauthorized access to certain cloud-based business applications, and that the incident was the result of a successful social engineering attack that compromised a user session associated with a third-party contractor. | | 2026-06-15 | [iRhythm Holdings, Inc.](/incidents/irhythm-hosted-apps-2026/) | not public | Stolen user credentials and session tokens | The filing says the affected data was obtained through social engineering from certain third-party-hosted business applications (the provider is not named). | | 2026-04-24 | [ADT Inc.](/incidents/adt-cloud-environments-2026/) | not public | Not public | The filing says the company became aware of unauthorized access to certain cloud-based environments, terminated the access, and notified law enforcement. How access was gained is not public. | | 2025-11-27 | [Mixpanel](/incidents/mixpanel-smishing-2025/) | Mixpanel | Stolen user credentials and session tokens | The company stated it detected a smishing campaign on November 8, 2025, and took steps to contain unauthorized access and secure impacted user accounts; it revoked sessions and rotated compromised credentials for impacted accounts. | | 2025-11-06 | [Illuminate Education, Inc.](/incidents/illuminate-education-2021/) | not public (the FTC says a third-party cloud provider, not named in the press release) | Retained access after departure or change | The FTC press release says that, according to the complaint, a hacker used the credentials of a former employee, who had left the company three and a half years earlier, to breach databases stored on a third-party cloud provider. It also says a third-party vendor had alerted the company in January 2020 to numerous security vulnerabilities and that student data was stored in plain text until at least January 2022. | | 2025-06-13 | [Zoomcar Holdings, Inc.](/incidents/zoomcar-cloud-2025/) | not public | Not public | The filing says the company identified unauthorized access to its information systems after employees received communications from a threat actor, and lists additional safeguards across the cloud and internal network as a response. How access was gained and which cloud provider was used are not public. | | 2025-05-01 | [The Walt Disney Company](/incidents/disney-slack-2024/) | Slack | Stolen user credentials and session tokens | The release says a malicious file posted online let an individual into an employee’s personal computer, where stored login credentials were obtained and used to reach the employee’s Slack account. | | 2025-04-28 | [Nova Scotia Power](/incidents/nova-scotia-power-opc-2025/) | not public | Stolen user credentials and session tokens | The letter says an employee visited a compromised website and clicked a link, malware was installed, the threat actor moved across systems using accounts with domain administrator privileges, harvested credentials, took data between 23 and 25 April 2025, then used harvested credentials to destroy backups and deploy ransomware. | | 2025-03-27 | [Advanced Computer Software Group Ltd (health and care subsidiary)](/incidents/advanced-computer-software-2022/) | not public | Stolen user credentials and session tokens | The ICO states hackers accessed certain systems of the company’s health and care subsidiary via a customer account that did not have multi-factor authentication (MFA). The ICO’s investigation found gaps in the deployment of MFA, a lack of comprehensive vulnerability scanning and inadequate patch management. The ICO does not state how the account was obtained. | | 2025-01-07 | [PowerSchool](/incidents/powerschool-powersource-2024/) | PowerSchool | Stolen user credentials and session tokens | The company stated, citing its forensic firm’s final report, that the threat actor accessed PowerSource, a community-focused customer support portal, using a single compromised credential. | | 2024-10-22 | [Avaya Holdings Corp.](/incidents/avaya-cloud-file-sharing-2020/) | not public | Third-party access | The SEC press release says the threat actor likely behind the SolarWinds Orion hack accessed the company’s systems without authorization. It describes the access as part of an investigation of public companies potentially impacted by the compromise of SolarWinds’ Orion software. The SEC charged the company over the accuracy of its public disclosure of the incident. | | 2024-08-30 | [Verkada Inc.](/incidents/verkada-2021/) | Verkada (cloud-based security camera provider) | Over-permissioned identities | The FTC press release says the complaint alleges the company failed to provide appropriate security measures, including failing to require unique and complex passwords, adequately encrypt customer data and implement secure network controls. The Justice Department release says the complaint alleges failures in appropriate access management and data protection controls. Neither source states how the hacker obtained the access. | | 2024-07-12 | [AT&T](/incidents/att-third-party-cloud-2024/) | not public | Not public | The filing says threat actors unlawfully accessed an AT&T workspace on a third-party cloud platform; the filing does not say how access was gained. | | 2024-06-14 | [Advance Auto Parts](/incidents/advance-auto-parts-2024/) | not public | Not public | The filing says the Company identified unauthorized activity within a third-party cloud database environment containing Company data; the filing does not say how access was gained. | | 2024-06-13 | [Pure Storage](/incidents/pure-storage-snowflake-workspace-2024/) | Snowflake | Not public | The filing says a third party temporarily gained unauthorized access to a single Snowflake data analytics workspace. How the access was gained is not public in the filing. | | 2024-05-31 | [Live Nation Entertainment](/incidents/live-nation-ticketmaster-2024/) | not public | Not public | The filing says the Company identified unauthorized activity within a third-party cloud database environment containing Company data; the filing does not say how access was gained. | | 2024-05-31 | [Hugging Face](/incidents/huggingface-spaces-secrets-2024/) | Hugging Face | Not public | The post says the company detected unauthorized access to its Spaces platform, specifically related to Spaces secrets. How access was gained is not public. The company stated it removed organization tokens, added key management for Spaces secrets and expanded detection of leaked tokens. | | 2024-05-30 | [Snowflake customer accounts (multiple organizations)](/incidents/snowflake-customer-accounts-2024/) | Snowflake | Stolen user credentials and session tokens | Snowflake stated in its annual report that a threat actor accessed a number of customers’ Snowflake accounts because those customers had not met their obligations to implement multi-factor authentication and network access policies, and that it found no breach of its own platform or environment. | | 2024-05-01 | [Dropbox](/incidents/dropbox-sign-2024/) | not public | Over-permissioned identities | The company stated a third party gained access to an automated system configuration tool, compromised a back-end service account with privileges across the production environment, and reached the customer database; it answered how access was gained with a compromised access token. | | 2024-01-24 | [Hewlett Packard Enterprise](/incidents/hpe-midnight-blizzard-email-2024/) | not public | Not public | The filing says a suspected nation-state actor, which the company believes to be the threat actor Midnight Blizzard, gained unauthorized access to HPE's cloud-based email environment. How access was gained is not public. | | 2023-12-11 | [A bank headquartered in San Francisco (not named in the U.S. Attorney release)](/incidents/sf-bank-cloud-engineer-retained-laptop-2020/) | not public | Retained access after departure or change | The release says a cloud engineer fired on March 11, 2020 used a company-issued laptop he had not returned to access the bank's computer network without authorization. | | 2023-11-16 | [Global Tel*Link Corp.](/incidents/global-tel-link-ftc-2020/) | not public | Public storage, databases and services | The FTC press release says that, after the company and a third-party vendor copied the data into the cloud for testing, changes made by the vendor to the security settings left the data accessible via the internet without safeguards, until a security researcher alerted the company. | | 2023-10-06 | [23andMe](/incidents/23andme-2023/) | not public | Stolen user credentials and session tokens | The company stated the threat actor accessed accounts through credential stuffing, using usernames and passwords that were also used on other websites previously compromised. | | 2023-08-30 | [Sourcegraph](/incidents/sourcegraph-2023/) | not public | Leaked access keys and secrets | The company stated a malicious actor used a leaked site-admin access token, accidentally committed in a pull request, to raise an account to site-admin on Sourcegraph.com. | | 2023-07-12 | [JumpCloud](/incidents/jumpcloud-orchestration-2023/) | JumpCloud | Stolen user credentials and session tokens | The company stated it traced anomalous activity on an internal orchestration system to a spear-phishing campaign on June 22, which gave unauthorized access to a specific area of its infrastructure. It later found unusual activity in the commands framework for a small set of customers, and stated the attack vector was data injection into its commands framework. | | 2023-03-10 | [Eatigo International Pte. Ltd.](/incidents/eatigo-pdpc-2020/) | not public | Public storage, databases and services | The decision says the legacy database sat on the infrastructure of a cloud service provider in Singapore, was left out of the company's private network setup, was reachable from the internet to anyone with the credentials, and had no password rotation rule. Staff no longer knew it existed, and no asset inventory or access logs were kept, so the company could not establish how or when access was gained. | | 2023-01-13 | [Mailchimp (Intuit)](/incidents/mailchimp-2023/) | Mailchimp | Stolen user credentials and session tokens | The company stated an unauthorized actor used a social engineering attack on employees and contractors and used the compromised employee credentials to reach an internal customer support and account administration tool. | | 2023-01-04 | [CircleCI](/incidents/circleci-2023/) | CircleCI | Stolen user credentials and session tokens | The company stated that malware on an engineer's laptop stole a valid 2FA-backed SSO session, and that the employee held privileges to generate production access tokens. | | 2022-12-22 | [LastPass](/incidents/lastpass-cloud-storage-2022/) | not public | Leaked access keys and secrets | The company stated that an unknown threat actor accessed a cloud-based storage environment using information obtained from the August 2022 incident, and that credentials and keys obtained from another targeted employee were used to access and decrypt some storage volumes. | | 2022-12-19 | [RedMart Limited](/incidents/redmart-pdpc-2020/) | AWS and Alibaba Cloud | Leaked access keys and secrets | The decision says a threat actor gained unauthorised access to the GitHub account of a member of the DevOps team, found a high-privilege key to the AWS environment in the repositories, used it to find a second key in a private S3 bucket, created an EC2 instance, and crossed a VPN link to the Alibaba cloud database. The database was neither encrypted nor password protected. | | 2022-12-06 | [Rackspace Technology, Inc.](/incidents/rackspace-hosted-exchange-2022/) | Rackspace Hosted Exchange | Not public | The filing does not say how access was gained. How access was gained is not public. | | 2022-10-31 | [Chegg](/incidents/chegg-ftc-2022/) | not public | Retained access after departure or change | The FTC stated that a former contractor used login information the company shared with employees and outside contractors to access one of Chegg's third-party cloud databases. | | 2022-10-19 | [Microsoft (misconfigured storage endpoint)](/incidents/microsoft-misconfigured-endpoint-2022/) | not public | Public storage, databases and services | The post-mortem says an unintentional misconfiguration on an endpoint, not a security vulnerability, allowed unauthenticated access; the endpoint is now accessible only with authentication. | | 2022-09-16 | [Uber](/incidents/uber-2022/) | Google Workspace | Stolen user credentials and session tokens | The company stated that an attacker likely bought a contractor's corporate password after malware on the contractor's personal device exposed it, then repeated two-factor login requests until the contractor accepted one. | | 2022-08-12 | [Mailchimp (Intuit)](/incidents/mailchimp-crypto-social-engineering-2022/) | Mailchimp | Stolen user credentials and session tokens | The company stated an unauthorized actor accessed an internal tool used by customer-facing teams for customer support and account administration, after a social engineering attack on Mailchimp employees that obtained employee credentials. | | 2022-08-10 | [Cisco Systems](/incidents/cisco-google-account-vishing-2022/) | Box | Stolen user credentials and session tokens | The company stated an employee's credentials were compromised after an attacker gained control of a personal Google account to which credentials saved in the browser were synchronised, and the attacker then used voice phishing and repeated multi-factor push requests until one was accepted, which gave access to the VPN. | | 2022-08-07 | [Twilio](/incidents/twilio-2022/) | Twilio | Stolen user credentials and session tokens | The company stated that text messages sent to current and former employees led some of them to a page impersonating the Twilio sign-in page, and that the attackers used the stolen credentials to reach internal systems. | | 2022-07-14 | [Quoine Pte Ltd](/incidents/quoine-pdpc-2020/) | not public | Third-party access | The decision says social engineering of employees of the company's domain registrar led to control of its domain hosting account being handed to an external actor, who redirected the company's mail, reset the password of a DevOps account, and used the reset credentials to reach the company's cloud computing platform, which held API keys to its databases. | | 2022-05-19 | [MGM Resorts International](/incidents/mgm-resorts-opc-2019/) | not public | Stolen user credentials and session tokens | The OPC finding, relaying MGM's account, says an unauthorized actor logged into a third-party platform used by MGM software developers with an MGM employee's credentials compromised in an earlier unrelated breach, exploited a coding flaw on that platform to bypass multi-factor authentication, found a temporary access token, and used it to reach guest data on an external cloud server for about one hour. | | 2022-03-22 | [Microsoft](/incidents/microsoft-dev-0537-2022/) | not public | Stolen user credentials and session tokens | The company stated that a single account was compromised, granting limited access, and that the tactics used in this intrusion reflect those described in the same post (including stolen credentials and MFA prompt abuse). | | 2021-11-22 | [GoDaddy](/incidents/godaddy-managed-wordpress-2021/) | GoDaddy Managed WordPress hosting | Stolen user credentials and session tokens | The company stated that an unauthorized third party used a compromised password to access the provisioning system in the legacy code base for Managed WordPress. | | 2021-10-14 | [Stylez Pte Ltd](/incidents/stylez-icompare-pdpc-2019/) | not public | Public storage, databases and services | The decision says the testing database was hosted on a cloud server leased from a cloud storage service provider, stored in a publicly accessible directory without access controls, so it was reachable from the internet and was crawled and indexed by search engines. | | 2021-08-30 | [Cetera Entities (Cetera Advisor Networks LLC, Cetera Investment Services LLC, Cetera Financial Specialists LLC, Cetera Advisors LLC, Cetera Investment Advisers LLC)](/incidents/cetera-email-account-takeovers-2017/) | not public | Stolen user credentials and session tokens | The SEC states the cloud-based email accounts of over 60 personnel were taken over by unauthorized third parties, and that none of the taken over accounts were protected in a manner consistent with the firms’ policies. The press release does not name the email provider or state how the accounts were taken over. | | 2021-08-30 | [Cambridge Investment Research Inc. and Cambridge Investment Research Advisors Inc.](/incidents/cambridge-email-account-takeovers-2018/) | not public | Stolen user credentials and session tokens | The SEC states the cloud-based email accounts of over 121 representatives were taken over by unauthorized third parties. The SEC order finds the firm discovered the first takeover in January 2018 but did not adopt and implement firm-wide enhanced security measures for those accounts until 2021. The press release does not name the email provider or state how the accounts were taken over. | | 2021-08-30 | [KMS Financial Services Inc.](/incidents/kms-financial-email-account-takeovers-2018/) | not public | Stolen user credentials and session tokens | The SEC states the cloud-based email accounts of 15 financial advisers or their assistants were taken over by unauthorized third parties. The SEC order finds the firm did not adopt written policies requiring additional firm-wide security measures until May 2020 and did not fully implement them firm-wide until August 2020. The press release does not name the email provider or state how the accounts were taken over. | | 2021-04-15 | [Codecov](/incidents/codecov-2021/) | Codecov | Leaked access keys and secrets | The company stated that an error in its Docker image creation process allowed the actor to extract the credential required to modify the Bash Uploader script, which was then altered to send CI environment variables to a third-party server. | | 2020-12-16 | [SkyMed International, Inc.](/incidents/skymed-ftc-2020/) | not public | Public storage, databases and services | The FTC press release says the database was unsecured and exposed by a security researcher, and that it could be located and accessed by anyone on the Internet. The press release says the company deleted the database after confirming that the data was online and publicly accessible. | | 2020-12-15 | [Ascension Data & Analytics](/incidents/ascension-data-2019/) | not public | Public storage, databases and services | The FTC complaint says a vendor hired for text recognition scanning stored the documents on a cloud-based server and a separate cloud-based storage location, misconfigured both, and left the information exposed to anyone on the internet; no password was required. The complaint says the respondent did not assess the vendor's security measures. | | 2020-10-16 | [MRI Diagnostics Pte Ltd and Clarity Radiology Pte Ltd](/incidents/mri-diagnostics-clarity-radiology-pdpc-2018/) | not public | Public storage, databases and services | The decision says the clinic subscribed to an internet-based teleradiology system provided by Clarity Radiology, a cloud-based software service, which engaged an overseas vendor to maintain it. The vendor removed access restrictions on a network folder to patch the system and did not restore them, so the spreadsheets were indexed by search engines. | | 2020-07-18 | [Twitter](/incidents/twitter-2020/) | Twitter | Stolen user credentials and session tokens | The company says a phone spear phishing attack targeted a small number of employees; the attackers used the employees' credentials to access internal systems, including getting through two-factor protections, then used the credentials of employees with access to account support tools. | | 2020-07-16 | [Blackbaud Inc.](/incidents/blackbaud-2020/) | Blackbaud (software provider; the New York Attorney General calls it a cloud company) | Over-permissioned identities | The FTC press release says that, according to the complaint, a hacker in early 2020 accessed a customer’s Blackbaud-hosted database and, once logged in, moved across multiple Blackbaud-hosted environments by using existing vulnerabilities and local administrator accounts and by creating new administrator accounts. The FTC says the company failed to segment data, adequately implement multifactor authentication and monitor for breach attempts. The company paid a ransom of 24 Bitcoin, according to the FTC. | ## Related pages - [AWS: cloud breaches in the ledger](/clouds/aws/) - [Amazon Web Services: cloud breaches in the ledger](/clouds/amazon-web-services/) - [GitHub: cloud breaches in the ledger](/clouds/github/) ## Sources - Veradigm Inc.: https://www.sec.gov/Archives/edgar/data/1124804/000119312526385249/mdrx-20260908.htm (read 16 Nov 2026) - Amgen: https://www.sec.gov/Archives/edgar/data/318154/000031815426000119/amgn-20260729.htm (read 16 Nov 2026) - AdaptHealth Corp.: https://www.sec.gov/Archives/edgar/data/1725255/000110465926080297/ahco-20260627x8k.htm (read 16 Nov 2026) - AdaptHealth: https://www.sec.gov/Archives/edgar/data/1725255/000110465926080297/ahco-20260627x8k.htm (read 16 Nov 2026) - iRhythm Holdings, Inc.: https://www.sec.gov/Archives/edgar/data/1388658/000138865826000055/irtc-20260610.htm (read 16 Nov 2026) - ADT Inc.: https://www.sec.gov/Archives/edgar/data/1703056/000170305626000038/adt-20260420.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/clouds/other/ --- # Cloud breaches involving AWS: 19 in the ledger Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. 19 incidents in the ledger involve AWS. The most common access path among them is leaked access keys and secrets. The earliest was disclosed in 2019 and the latest in 2026. Each incident links to its company statement, filing, court document or post-mortem. | Disclosed | Organization | Access path | What was exposed | | --- | --- | --- | --- | | 2026-01-08 | [People Central Pte. Ltd.](/incidents/people-central-pdpc-2024/) | Exposed network services and request paths | The PDPC decision states personal data of 95,000 employees of the company's clients was put at risk (including name, identity card number, salary, bank account number and address), plus 24,765 emergency contacts and children. The threat actor deleted databases and likely exfiltrated data, and data was found for sale on the dark web. | | 2025-08-20 | [Salesloft (Drift)](/incidents/salesloft-drift-2025/) | Leaked access keys and secrets | OAuth tokens for Drift customers' technology integrations, obtained from the Drift AWS environment; Salesloft GitHub repositories downloaded; environment variable secrets exfiltrated | | 2025-07-23 | [Amazon Web Services](/incidents/aws-amazon-q-extension-2025/) | Over-permissioned identities | Malicious code was committed to the open-source repository of the Amazon Q Developer for VS Code extension and distributed in version 1.84.0; AWS stated it was unsuccessful in executing and made no changes to services or customer environments | | 2024-04-22 | [Payroll2U Pte. Ltd.](/incidents/payroll2u-pdpc-2022/) | Stolen user credentials and session tokens | The PDPC decision states 81.95 GB of data was exfiltrated and posted on the dark web, including the personal data of 5,640 employees of a client: name, bank account number, salary, identity card number, address, date of birth and email. | | 2023-11-07 | [Sumo Logic](/incidents/sumo-logic-aws-credential-2023/) | Leaked access keys and secrets | The company stated that, after investigation by its teams and third-party forensic experts, it found no proof of customer data impact. It advised customers to rotate API access keys and stored third-party credentials. | | 2023-08-16 | [E-Commerce Enablers Pte. Ltd.](/incidents/ecommerce-enablers-pdpc-2020/) | Leaked access keys and secrets | The PDPC decision states customer data on virtual servers in an AWS environment was exfiltrated, with counts including 1,457,637 email addresses, 840,210 names, 447,076 mobile numbers, 299,381 bank account numbers and 378,531 partial credit card records. | | 2023-06-16 | [1Health.io Inc. (formerly Vitagene, Inc.)](/incidents/vitagene-1health-ftc-2019/) | Public storage, databases and services | The FTC states nearly 2,400 health reports about consumers and raw genetic data of at least 227 consumers, sometimes accompanied by a first name, sat in publicly accessible buckets, unencrypted, according to the complaint. | | 2022-10-24 | [Drizly](/incidents/drizly-ftc-2022/) | Stolen user credentials and session tokens | The FTC complaint stated that Drizly's databases held personal information of more than 2.5 million consumers, that the intruder took a user table of more than 2.5 million records, and that the company hosted its production database on Amazon RDS, an AWS service. | | 2022-09-15 | [MyRepublic Limited](/incidents/myrepublic-pdpc-2021/) | Leaked access keys and secrets | The PDPC decision states the personal data of 79,388 customers was accessed and exfiltrated: scanned copies of identity cards and work passes submitted as KYC documents and number portability documents. | | 2022-07-14 | [Crawfort Pte. Ltd.](/incidents/crawfort-pdpc-2020/) | Public storage, databases and services | The PDPC decision states the personal data of 5,421 customers was affected: images of identity cards, PDF loan contracts and income documents. | | 2021-12-01 | [A publicly traded wireless communications manufacturer (called Company-1 in the U.S. Attorney release)](/incidents/ubiquiti-aws-github-2020/) | Over-permissioned identities | The release says gigabytes of confidential data were downloaded from the company’s AWS and GitHub infrastructure and later used in an extortion attempt for millions of dollars. | | 2021-11-11 | [Commeasure Pte Ltd (operates the RedDoorz booking platform)](/incidents/commeasure-reddoorz-pdpc-2020/) | Leaked access keys and secrets | The PDPC decision states a database of 5,892,843 customer records (name, contact number, email, date of birth, hashed password and booking information) was accessed and exfiltrated; masked credit card numbers were not accessed. | | 2021-09-21 | [Sendtech Pte. Ltd.](/incidents/sendtech-pdpc-2021/) | Leaked access keys and secrets | The PDPC decision states the personal data of 64,196 customers and 3,401 contractors and the contractors' employees was accessed: for customers the email address, contact number, home address and last four digits of the card; for contractors and their employees a profile photo and copies of the NRIC or work permit. It states there was no evidence of data exfiltration. | | 2020-08-26 | [Cisco Systems](/incidents/cisco-webex-aws-2018/) | Retained access after departure or change | The press release says 456 virtual machines for the WebEx Teams application were deleted and over 16,000 WebEx Teams accounts were shut down for up to two weeks. It says the company spent approximately $1,400,000 in employee time and refunded over $1,000,000 to affected customers, and that no customer data was compromised. | | 2020-07-22 | [Twilio](/incidents/twilio-2020/) | Public storage, databases and services | A modified version (v1.20) of the TaskRouter JS SDK hosted on the company's content domain. The company stated it had no evidence that any customer data was accessed and that no malicious party had access to its internal systems, code or data; customers who downloaded v1.20 in the window were asked to replace it. | | 2020-07-11 | [LiveAuctioneers](/incidents/liveauctioneers-2020/) | Third-party access | The company's letter says 38,523 Washington residents were involved; the user database held names, email addresses (user names), MD5-hashed passwords, contact details, IP addresses used to create accounts and records of past sales transactions. It says no Social Security numbers or financial data were collected. | | 2019-08-27 | [Imperva](/incidents/imperva-2019/) | Leaked access keys and secrets | Elements of the Incapsula (Cloud WAF) customer database through September 15, 2017: email addresses and hashed and salted passwords, and for a subset of customers API keys and customer-provided SSL certificates. The company stated the exposure came from a database snapshot. | | 2019-07-29 | [Capital One](/incidents/capital-one-2019/) | Exposed network services and request paths | The company stated the event affected approximately 100 million individuals in the United States and approximately 6 million in Canada. Credit card application data from 2005 through early 2019 (names, addresses, ZIP or postal codes, phone numbers, email addresses, dates of birth, self-reported income), portions of credit card customer data, about 140,000 Social Security numbers and about 80,000 linked bank account numbers of US secured credit card customers, and approximately 1 million Canadian Social Insurance Numbers. The company stated no credit card account numbers or log-in credentials were compromised. | | 2019-06-28 | [Online Buddies (Jack’d dating app)](/incidents/online-buddies-jackd-s3-2018/) | Public storage, databases and services | The release says private photos (which may have included nude images), public photos and personal information (location, device ID, last login date, hashed password) were at risk of unauthorized access; counts not public. | ## How access was gained, entry by entry - [People Central Pte. Ltd.](/incidents/people-central-pdpc-2024/): The decision says the company, a cloud-based HR software provider, had Remote Desktop Protocol access open to the internet without two-factor authentication on its AWS cloud servers, settings allowed inbound traffic from all IP addresses in the cloud environment, and SQL injection vulnerabilities were present with no web application firewall. - [Salesloft (Drift)](/incidents/salesloft-drift-2025/): The investigation summary says the threat actor accessed the Salesloft GitHub account (March to June 2025), then accessed Drift's AWS environment and obtained OAuth tokens for customers' integrations, and used them to access data via Drift integrations. - [Amazon Web Services](/incidents/aws-amazon-q-extension-2025/): The bulletin says the extension had an inappropriately scoped GitHub token in its CodeBuild configuration, and with that token the threat actor committed malicious code. - [Payroll2U Pte. Ltd.](/incidents/payroll2u-pdpc-2022/): The decision says a single compromised user account was used for Remote Desktop Protocol access to five servers in the company's AWS environment, and that investigators could not determine how the credentials were obtained. It lists no MFA for administrator accounts or accounts with access to personal data. - [Sumo Logic](/incidents/sumo-logic-aws-credential-2023/): The customer notice says the activity used a compromised credential to access a Sumo Logic AWS account. How the credential was compromised is not public. - [E-Commerce Enablers Pte. Ltd.](/incidents/ecommerce-enablers-pdpc-2020/): The decision says an AWS access key with full administrative privileges was committed to a private GitHub repository on 4 June 2019, removed on 6 June but still viewable in the commit history, and not fully disabled when a replacement key was created. On 9 September 2020 a threat actor used it to reach the AWS environment, changed security settings to allow remote internet access to database instances and exfiltrated data. The company conjectured the key was found in the commit history. - [1Health.io Inc. (formerly Vitagene, Inc.)](/incidents/vitagene-1health-ftc-2019/): The FTC press release says the company stored the data in publicly accessible buckets on AWS cloud storage, did not encrypt it, restrict access to it, log or monitor access to it, or inventory it, and that it was warned at least three times over a two-year period before investigating after a security researcher contacted it in June 2019. - [Drizly](/incidents/drizly-ftc-2022/): The FTC complaint stated that in early July 2020 a malicious actor reused credentials from an unrelated breach to reach an executive's GitHub account, found AWS and database credentials stored in the company's GitHub repositories, and used them to change the company's AWS security settings and reach the production environment. - [MyRepublic Limited](/incidents/myrepublic-pdpc-2021/): The decision says the Mobile Order Portal stored KYC documents in a bucket on AWS cloud storage, and the bucket's access key sat in the portal's source code. The external actor used the key to access the bucket. The company could not determine with certainty how the key was obtained but found it was disclosed in a portal page showing technical information and in a source repository open to all its developers. - [Crawfort Pte. Ltd.](/incidents/crawfort-pdpc-2020/): The decision says the company opened an S3 server port in its AWS environment for one week during a data migration, and that the S3 bucket holding the data was publicly accessible due to a misconfiguration. It states the threat actor gained access to the public bucket in that period. - [A publicly traded wireless communications manufacturer (called Company-1 in the U.S. Attorney release)](/incidents/ubiquiti-aws-github-2020/): The release says an employee who was a senior developer held credentials for the company’s AWS and GitHub servers and repeatedly misused his administrative access to download the data; the company’s own account of an outside intruder was, the release says, false. - [Commeasure Pte Ltd (operates the RedDoorz booking platform)](/incidents/commeasure-reddoorz-pdpc-2020/): The decision says the threat actors most likely obtained an AWS access key embedded in an Android application package, created in 2015, last updated in January 2018, still downloadable from the Google Play Store, and marked as a test key although it could reach a live production database in Amazon RDS. - [Sendtech Pte. Ltd.](/incidents/sendtech-pdpc-2021/): The decision says the cause was a compromised AWS access key created in 2015 and not rotated since. With it the attacker gained admin privileges, created another admin account and queried the buckets storing personal data. The company suspected the key could have been compromised through former or current employees but the decision does not confirm how. - [Cisco Systems](/incidents/cisco-webex-aws-2018/): The Department of Justice says that under the plea agreement an individual who had worked for Cisco and resigned in approximately April 2018 admitted intentionally accessing Cisco's cloud infrastructure hosted by Amazon Web Services without permission on September 24, 2018, and deployed code from a Google Cloud Project account that deleted the virtual machines. - [Twilio](/incidents/twilio-2020/): The company's incident report says a misconfiguration in the S3 bucket hosting the library let a bad actor inject code, and that the access policy for the path storing the SDK allowed anybody on the Internet to read and write to that path. - [LiveAuctioneers](/incidents/liveauctioneers-2020/): The company's letter to the Washington Attorney General says a hacker breached one of its third-party software solutions and obtained a source control key, which gave access to its source repository, its cloud-based storage environment and its production database, then used internal users' credentials to download the user database. - [Imperva](/incidents/imperva-2019/): The company's follow-up post says an unauthorized use of an administrative API key in one of its production AWS accounts led to exposure of a database snapshot. It lists the steps as a snapshot created for testing, an internal compute instance accessible from the outside world that contained an AWS API key, compromise of that instance and theft of the key, and use of the key to access the snapshot. - [Capital One](/incidents/capital-one-2019/): The company stated that a highly sophisticated individual was able to exploit a specific configuration vulnerability in its infrastructure and that the issue was not specific to the cloud. The OCC consent order finds the bank failed to establish appropriate risk management for its cloud operating environment, including network security controls. - [Online Buddies (Jack’d dating app)](/incidents/online-buddies-jackd-s3-2018/): The New York Attorney General’s investigation confirmed the company failed to secure user photos stored in an AWS S3 storage service, and left app interfaces to backend data unsecured. ## Related pages - [Amazon Web Services: cloud breaches in the ledger](/clouds/amazon-web-services/) - [GitHub: cloud breaches in the ledger](/clouds/github/) - [Microsoft 365: cloud breaches in the ledger](/clouds/microsoft-365/) ## Sources - People Central Pte. Ltd.: https://www.pdpc.gov.sg/assets/f9fdd00e-2e3d-4e69-a017-4da0db04d1ad (read 16 Nov 2026) - Salesloft (Drift): https://trust.salesloft.com/?uid=Drift+Security+Update (read 16 Nov 2026) - Amazon Web Services: https://aws.amazon.com/security/security-bulletins/AWS-2025-015/ (read 16 Nov 2026) - Payroll2U Pte. Ltd.: https://www.pdpc.gov.sg/assets/f3d9bace-cb3d-4e01-ab1c-5099c882744e (read 16 Nov 2026) - Sumo Logic: https://web.archive.org/web/20240101000000/https://www.sumologic.com/security-response-center/ (read 16 Nov 2026) - E-Commerce Enablers Pte. Ltd.: https://www.pdpc.gov.sg/assets/777b18ae-71e5-4e5d-9e89-9d1b54aca926 (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/clouds/aws/ --- # Cloud breaches involving Amazon Web Services: 3 in the ledger Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. 3 incidents in the ledger involve Amazon Web Services. The most common access path among them is exposed network services and request paths. The earliest was disclosed in 2022 and the latest in 2025. Each incident links to its company statement, filing, court document or post-mortem. | Disclosed | Organization | Access path | What was exposed | | --- | --- | --- | --- | | 2025-03-20 | [People Central Pte. Ltd. (Singapore, cloud-based HR software provider)](/incidents/people-central-aws-rdp-2024/) | Exposed network services and request paths | The decision says a threat actor deleted databases and likely exfiltrated data from the company's AWS servers; data of 95,000 employees of its clients (including NRIC number, salary and bank account number) and of 24,765 emergency contacts and children was put at risk, and some data was found for sale on the dark web. | | 2024-12-08 | [BeyondTrust](/incidents/beyondtrust-remote-support-saas-2024/) | Exposed network services and request paths | The company stated a security incident involved 17 Remote Support SaaS customers: an infrastructure API key was used to enable access to certain Remote Support SaaS instances by resetting local application passwords. It stated no other BeyondTrust products or systems were affected and ransomware was not involved. | | 2022-02-23 | [Alfi, Inc.](/incidents/alfi-former-executives-cloud-lockout-2022/) | Retained access after departure or change | The filing says the company's Google Workspace (email accounts, calendars, contacts, chats and saved files) and its Amazon Web Services account (which hosts tablet data and a software development platform) were taken over so the company could not administer them, that downloadable access to all data in the Google Workspace was obtained from Google, and that files were accessed and downloaded from the company's Google Drive. The filing says the company continued business through workarounds. | ## How access was gained, entry by entry - [People Central Pte. Ltd. (Singapore, cloud-based HR software provider)](/incidents/people-central-aws-rdp-2024/): The decision lists contributing lapses: SQL injection vulnerabilities in the web application with no web application firewall, RDP access open to the internet without two-factor authentication, settings that allowed inbound traffic from all IP addresses in the cloud environment, and vulnerability scanning only every two years. - [BeyondTrust](/incidents/beyondtrust-remote-support-saas-2024/): The company stated a zero-day vulnerability of a third-party application was used to gain access to an online asset in a BeyondTrust AWS account, which allowed the threat actor to obtain an infrastructure API key that was then used against a separate AWS account that operated Remote Support infrastructure. - [Alfi, Inc.](/incidents/alfi-former-executives-cloud-lockout-2022/): The filing says that after the executives were placed on leave and told not to access the company's computer systems or platforms, one of them took control of the Google Workspace and Amazon Web Services accounts, and two others accessed and downloaded files from Google Drive. ## Related pages - [GitHub: cloud breaches in the ledger](/clouds/github/) - [Microsoft 365: cloud breaches in the ledger](/clouds/microsoft-365/) - [Microsoft Azure: cloud breaches in the ledger](/clouds/microsoft-azure/) ## Sources - People Central Pte. Ltd. (Singapore, cloud-based HR software provider): https://www.pdpc.gov.sg/assets/f9fdd00e-2e3d-4e69-a017-4da0db04d1ad?v=1 (read 16 Nov 2026) - BeyondTrust: https://www.beyondtrust.com/remote-support-saas-service-security-investigation (read 16 Nov 2026) - Alfi, Inc.: https://www.sec.gov/Archives/edgar/data/1833908/000110465922026150/tm227500d1_8k.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/clouds/amazon-web-services/ --- # Cloud breaches involving GitHub: 9 in the ledger Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. 9 incidents in the ledger involve GitHub. The most common access path among them is stolen user credentials and session tokens. The earliest was disclosed in 2022 and the latest in 2026. Each incident links to its company statement, filing, court document or post-mortem. | Disclosed | Organization | Access path | What was exposed | | --- | --- | --- | --- | | 2026-05-08 | [SailPoint, Inc.](/incidents/sailpoint-github-2026/) | Third-party access | A subset of the company’s GitHub repositories; the filing says no evidence that customer data in production or staging environments was accessed, and customers with information in the accessed repositories were notified. | | 2025-09-22 | [GitHub (npm registry)](/incidents/github-npm-shai-hulud-2025/) | Stolen user credentials and session tokens | The post says 500+ compromised packages were removed from the npm registry and that the worm could steal multiple types of secrets, not just npm tokens. | | 2025-08-26 | [Nx (Nrwl)](/incidents/nx-github-actions-2025/) | Leaked access keys and secrets | The company's npm publishing token; malicious package versions scanned users' systems and uploaded results to public GitHub repositories. The company stated Nx Cloud was not affected. | | 2023-03-23 | [GitHub](/incidents/github-rsa-ssh-key-2023/) | Leaked access keys and secrets | The company stated GitHub.com's RSA SSH private key was briefly exposed in a public GitHub repository; it stated it had no reason to believe the key was abused and that no GitHub systems or customer information were compromised. | | 2022-12-31 | [Slack](/incidents/slack-github-tokens-2022/) | Stolen user credentials and session tokens | The company stated that private code repositories were downloaded, that they contained no customer data, means to access customer data or Slack's primary codebase, and that the production environment was not accessed. | | 2022-11-01 | [Dropbox](/incidents/dropbox-github-phishing-2022/) | Stolen user credentials and session tokens | The company stated that the threat actor copied 130 code repositories containing some credentials (primarily API keys) and a few thousand names and email addresses; no customer content, passwords or payment information was accessed. | | 2022-09-16 | [GitHub users (phishing campaign reported by GitHub)](/incidents/github-circleci-phishing-2022/) | Stolen user credentials and session tokens | GitHub stated that the campaign affected many victim organizations, and that in many cases the threat actor downloaded private repository contents accessible to the compromised user; GitHub itself was not affected. | | 2022-04-21 | [GeniusU Pte. Ltd.](/incidents/geniusu-pdpc-2021/) | Leaked access keys and secrets | The PDPC decision states about 1.26 million users' first and last name, email address, location and last sign-in IP address were accessed and exfiltrated from a staging application database. | | 2022-04-15 | [Heroku (Salesforce) and Travis CI OAuth tokens](/incidents/heroku-travisci-oauth-2022/) | Third-party access | GitHub stated that stolen OAuth user tokens issued to Heroku and Travis CI were used to download data from dozens of organizations, including npm; Heroku stated the actor also downloaded stored customer GitHub integration tokens and a database of pipeline-level config variables. | ## How access was gained, entry by entry - [SailPoint, Inc.](/incidents/sailpoint-github-2026/): The filing says the company detected unauthorized access to a subset of its GitHub repositories and that the root cause was a vulnerability in a third-party application, which it does not name. - [GitHub (npm registry)](/incidents/github-npm-shai-hulud-2025/): GitHub stated that a self-replicating worm spread through compromised maintainer accounts by injecting malicious post-install scripts into popular packages. - [Nx (Nrwl)](/incidents/nx-github-actions-2025/): The post-mortem says attackers exploited a GitHub Actions injection vulnerability to steal the npm publishing token; the repository's workflow permissions were read and write. - [GitHub](/incidents/github-rsa-ssh-key-2023/): The company stated the exposure resulted from what it believes was an inadvertent publishing of private information. - [Slack](/incidents/slack-github-tokens-2022/): The company stated that a limited number of employee tokens were stolen and misused to reach its externally hosted GitHub repository, and that a third-party vendor was compromised. - [Dropbox](/incidents/dropbox-github-phishing-2022/): The company stated that employees received phishing emails impersonating a CI service, and that a fake login page collected GitHub credentials and a one-time code, giving the threat actor access to one GitHub organization. - [GitHub users (phishing campaign reported by GitHub)](/incidents/github-circleci-phishing-2022/): GitHub stated that a phishing site imitating the GitHub login page stole credentials and relayed TOTP codes to the threat actor in real time, and that accounts protected by hardware security keys were not vulnerable. - [GeniusU Pte. Ltd.](/incidents/geniusu-pdpc-2021/): The decision says the likely cause was a compromise of a developer's GitHub password, which let the threat actor into the company's GitHub environment; database login credentials had been stored in the codebase there, so the actor reached and exfiltrated the database. The staging database held real personal data. - [Heroku (Salesforce) and Travis CI OAuth tokens](/incidents/heroku-travisci-oauth-2022/): Heroku stated that initial access to its environment was obtained by using a compromised token for a Heroku machine account, after which stored customer GitHub integration OAuth tokens were downloaded; GitHub stated the tokens were then abused against GitHub.com. ## Related pages - [Microsoft 365: cloud breaches in the ledger](/clouds/microsoft-365/) - [Microsoft Azure: cloud breaches in the ledger](/clouds/microsoft-azure/) - [Okta: cloud breaches in the ledger](/clouds/okta/) ## Sources - SailPoint, Inc.: https://www.sec.gov/Archives/edgar/data/2030781/000162828026033016/sail-20260508.htm (read 16 Nov 2026) - GitHub (npm registry): https://github.blog/security/supply-chain-security/our-plan-for-a-more-secure-npm-supply-chain/ (read 16 Nov 2026) - Nx (Nrwl): https://nx.dev/blog/s1ngularity-postmortem (read 16 Nov 2026) - GitHub: https://github.blog/2023-03-23-we-updated-our-rsa-ssh-host-key/ (read 16 Nov 2026) - Slack: https://slack.com/blog/news/slack-security-update (read 16 Nov 2026) - Dropbox: https://dropbox.tech/security/a-recent-phishing-campaign-targeting-dropbox (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/clouds/github/ --- # Cloud breaches involving Microsoft 365: 9 in the ledger Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. 9 incidents in the ledger involve Microsoft 365. The most common access path among them is stolen user credentials and session tokens. The earliest was disclosed in 2020 and the latest in 2026. Each incident links to its company statement, filing, court document or post-mortem. | Disclosed | Organization | Access path | What was exposed | | --- | --- | --- | --- | | 2026-08-06 | [IEH Corporation](/incidents/ieh-microsoft-365-2026/) | Stolen user credentials and session tokens | Mailbox contents of one employee, including email messages, attachments, customer communications, purchase orders and engineering-related documentation; the filing says no evidence currently exists that data was exfiltrated | | 2026-05-01 | [A Maryland medical system (called Company A in the indictment release)](/incidents/maryland-medical-system-m365-2016/) | Stolen user credentials and session tokens | The release says usernames, passwords, cookies, images, videos and other data of 195 victims, who were current or former employees or people affiliated with them, were obtained. | | 2024-01-19 | [Microsoft](/incidents/microsoft-midnight-blizzard-2024/) | Stolen user credentials and session tokens | Microsoft stated that some emails and attached documents were exfiltrated from a very small percentage of corporate email accounts, including senior leadership and cybersecurity and legal staff; a March 2024 update says some source code repositories and internal systems were also accessed. | | 2023-07-11 | [Microsoft](/incidents/microsoft-storm-0558-2023/) | Cross-tenant access | Microsoft stated that email accounts affecting approximately 25 organizations in the public cloud, including government agencies, and related consumer accounts were accessed. | | 2022-11-18 | [Farrer Park Hospital Pte Ltd](/incidents/farrer-park-hospital-pdpc-2018/) | Not public | The PDPC decision states 9,271 emails were automatically forwarded from two employees' Microsoft Office 365 mailboxes to a third-party address, disclosing the personal data of 3,539 individuals, 1,923 of them with medical information. | | 2021-10-14 | [The National Kidney Foundation](/incidents/national-kidney-foundation-pdpc-2020/) | Stolen user credentials and session tokens | The PDPC decision states the mailbox of one employee held 23,145 emails with the personal data of approximately 500 individuals (patients, employees and third parties), including bank account numbers, dialysis readings and medical conditions. It states the mailbox was believed to have been downloaded in full. | | 2021-03-16 | [Mimecast](/incidents/mimecast-solarwinds-certificate-2021/) | Third-party access | The company stated the threat actor accessed certain Mimecast-issued certificates and related customer server connection information, a subset of email addresses and other contact information, hashed and salted credentials, and a limited number of source code repositories. It stated Microsoft informed it that a certificate used to authenticate some Mimecast products to Microsoft 365 Exchange Web Services was used to connect to a low single-digit number of mutual customers' Microsoft 365 tenants. It stated it had no evidence that email or archive content held for customers was accessed. | | 2020-12-14 | [SolarWinds](/incidents/solarwinds-2020/) | Third-party access | The filing says a vulnerability was inserted in Orion monitoring product updates; fewer than 18,000 of approximately 33,000 notified Orion customers may have had an installation containing it. The filing also says the company's own Microsoft Office 365 email was compromised and that it had uncovered no evidence at that time of exfiltration of customer, personnel or other data. The Microsoft post describes actions in the Microsoft cloud against customer tenants. | | 2020-09-10 | [Singapore Medical Association](/incidents/singapore-medical-association-pdpc-2020/) | Stolen user credentials and session tokens | The PDPC decision states the personal data of 68 individuals in 137 emails (identity card numbers, dates of birth, indemnity coverage details, education and financial transaction information) was forwarded to an external address. | ## How access was gained, entry by entry - [IEH Corporation](/incidents/ieh-microsoft-365-2026/): The filing says a phishing email carried a link disguised as a Microsoft document-sharing link, and the user entered Microsoft 365 credentials into a fraudulent login page. - [A Maryland medical system (called Company A in the indictment release)](/incidents/maryland-medical-system-m365-2016/): The release says an employee used keylogging, cookie managers and mailbox rules to reach victims’ personal and professional accounts, including Microsoft 365, and exported browser cookies to keep unauthorized access from other devices. - [Microsoft](/incidents/microsoft-midnight-blizzard-2024/): Microsoft stated the threat actor used a password spray attack to compromise a legacy non-production test tenant account, then used that account's permissions to access corporate email accounts. - [Microsoft](/incidents/microsoft-storm-0558-2023/): Microsoft stated the actor used an acquired Microsoft account consumer signing key to forge tokens, and exploited a token validation issue so that enterprise mail accepted them. The post-mortem says the most probable mechanism for the key acquisition was a compromised engineer corporate account with access to a debugging environment holding a crash dump that contained the key. - [Farrer Park Hospital Pte Ltd](/incidents/farrer-park-hospital-pdpc-2018/): The decision says the two mailboxes had been configured to forward all incoming emails to a third party. It was found on 24 October 2019 when Office 365 restricted the accounts, which the decision describes as a security feature indicating unauthorised access. The decision does not state how the forwarding was set up; multi-factor authentication was not in place when it began in March 2018. - [The National Kidney Foundation](/incidents/national-kidney-foundation-pdpc-2020/): The decision says an employee received a phishing email linking to a website that asked for account credentials; the hacker is believed to have obtained them, accessed the employee's mailbox, which was hosted on Microsoft Office 365 and reachable by webmail, synchronised it on 17 May 2020, and used it to send phishing emails to 1,039 external contacts. - [Mimecast](/incidents/mimecast-solarwinds-certificate-2021/): The company stated the threat actor used the SolarWinds supply-chain compromise to gain access to part of its production grid environment, and from there accessed Mimecast-issued certificates and related connection information; the certificate was then used against a small number of customer Microsoft 365 tenants. - [SolarWinds](/incidents/solarwinds-2020/): The filing says the vulnerability was introduced as a result of a compromise of the Orion software build system, and that the company was made aware of an attack vector used to compromise its Office 365 emails. The Microsoft post says that in actions observed in the Microsoft cloud, attackers gained administrative access either using compromised privileged account credentials (for example stolen passwords) or by forging SAML tokens using compromised token signing certificates. - [Singapore Medical Association](/incidents/singapore-medical-association-pdpc-2020/): The decision says the association believed an unauthorised user entered a Microsoft Office 365 email account by a brute force attack but had no logs to confirm, then created a rule forwarding received emails to an external address. The account password had not been changed since November 2013 and there were no measures against repeated login attempts. ## Related pages - [Microsoft Azure: cloud breaches in the ledger](/clouds/microsoft-azure/) - [Okta: cloud breaches in the ledger](/clouds/okta/) - [Salesforce: cloud breaches in the ledger](/clouds/salesforce/) ## Sources - IEH Corporation: https://www.sec.gov/Archives/edgar/data/50292/000117494726000761/form8k-36187_iehc.htm (read 16 Nov 2026) - A Maryland medical system (called Company A in the indictment release): https://web.archive.org/web/2026/https://www.justice.gov/usao-md/pr/maryland-man-indicted-unauthorized-computer-access-related-maryland-medical-system-and (read 16 Nov 2026) - Microsoft: https://www.microsoft.com/en-us/msrc/blog/2024/01/microsoft-actions-following-attack-by-nation-state-actor-midnight-blizzard (read 16 Nov 2026) - Microsoft: https://www.microsoft.com/en-us/msrc/blog/2023/07/microsoft-mitigates-china-based-threat-actor-storm-0558-targeting-of-customer-email (read 16 Nov 2026) - Farrer Park Hospital Pte Ltd: https://www.pdpc.gov.sg/assets/685c6f15-4fd9-49be-88e4-b86975ce7974 (read 16 Nov 2026) - The National Kidney Foundation: https://www.pdpc.gov.sg/assets/8238cce1-16fd-4cec-b9ab-ffd8e8445a7b (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/clouds/microsoft-365/ --- # Cloud breaches involving Microsoft Azure: 5 in the ledger Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. 5 incidents in the ledger involve Microsoft Azure. The most common access path among them is public storage, databases and services. The earliest was disclosed in 2020 and the latest in 2023. Each incident links to its company statement, filing, court document or post-mortem. | Disclosed | Organization | Access path | What was exposed | | --- | --- | --- | --- | | 2023-09-18 | [Microsoft](/incidents/microsoft-ai-sas-token-2023/) | Leaked access keys and secrets | Microsoft stated that data in an internal storage account included backups of two former employees' workstation profiles and internal Microsoft Teams messages of those employees; it stated no customer data was exposed. | | 2022-06-30 | [Wegmans](/incidents/wegmans-azure-containers-2021/) | Public storage, databases and services | The release says a database backup with over three million records of customer email addresses and account passwords sat in the first container; a second container held a database with customers’ names, email addresses and mailing addresses; the release puts the total at more than three million consumers, including more than 830,000 New Yorkers. | | 2022-06-30 | [Wegmans Food Markets, Inc.](/incidents/wegmans-azure-container-2018/) | Public storage, databases and services | The New York Attorney General states the data of more than three million consumers nationwide, including more than 830,000 New Yorkers, was exposed, including a database backup file with over three million records of customer email addresses and account passwords. | | 2021-08-27 | [Microsoft (Azure Cosmos DB)](/incidents/azure-cosmosdb-chaosdb-2021/) | Cross-tenant access | The post-mortem says primary read-write keys of a subset of Azure Cosmos DB customers could be reached, customers were notified to regenerate keys, and no customer data was accessed by third parties or security researchers. | | 2020-01-22 | [Microsoft](/incidents/microsoft-support-database-2019/) | Public storage, databases and services | An internal customer support database used for support case analytics. The post says most customers did not have personally identifiable information exposed, that records are redacted with automated tools, and that some data may have remained unredacted, for example an email address written in a non-standard format. Counts: not public. | ## How access was gained, entry by entry - [Microsoft](/incidents/microsoft-ai-sas-token-2023/): Microsoft stated an employee shared a URL for a blob store in a public GitHub repository, and the URL included an overly-permissive Shared Access Signature token for an internal storage account. - [Wegmans](/incidents/wegmans-azure-containers-2021/): A security researcher told the company in April 2021 that a cloud storage container hosted on Microsoft Azure was open to public access; the release says a second container had been public since November 2018. - [Wegmans Food Markets, Inc.](/incidents/wegmans-azure-container-2018/): The New York Attorney General states that in April 2021 a security researcher informed the company that a cloud storage container hosted on Microsoft Azure was left unsecured and open to public access. The Attorney General’s office found the company failed to configure the container to limit access, to inventory its cloud assets, to secure all user passwords and to regularly conduct security testing of its cloud assets. - [Microsoft (Azure Cosmos DB)](/incidents/azure-cosmosdb-chaosdb-2021/): The post-mortem says a vulnerability in the Azure Cosmos DB Jupyter Notebook feature could allow a user to gain access to another customer's resources by using the account's primary read-write key. - [Microsoft](/incidents/microsoft-support-database-2019/): The post says a change made to the database's network security group on December 5, 2019 contained misconfigured security rules that enabled exposure of the data. It says the investigation found no malicious use, and that the issue was specific to an internal database and does not represent an exposure of the company's commercial cloud services. ## Related pages - [Okta: cloud breaches in the ledger](/clouds/okta/) - [Salesforce: cloud breaches in the ledger](/clouds/salesforce/) - [Cloud breaches: other providers and provider not public](/clouds/other/) ## Sources - Microsoft: https://www.microsoft.com/en-us/msrc/blog/2023/09/microsoft-mitigated-exposure-of-internal-information-in-a-storage-account-due-to-overly-permissive-sas-token (read 16 Nov 2026) - Wegmans: https://ag.ny.gov/press-release/2022/attorney-general-james-secures-400000-wegmans-after-data-breach-exposed-consumers (read 16 Nov 2026) - Wegmans Food Markets, Inc.: https://ag.ny.gov/press-release/2022/attorney-general-james-secures-400000-wegmans-after-data-breach-exposed-consumers (read 16 Nov 2026) - Microsoft (Azure Cosmos DB): https://web.archive.org/web/2021/https://msrc-blog.microsoft.com/2021/08/27/update-on-vulnerability-in-the-azure-cosmos-db-jupyter-notebook-feature/ (read 16 Nov 2026) - Microsoft: https://www.microsoft.com/en-us/msrc/blog/2020/01/access-misconfiguration-for-customer-support-database (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/clouds/microsoft-azure/ --- # Cloud breaches involving Okta: 7 in the ledger Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. 7 incidents in the ledger involve Okta. The most common access path among them is third-party access. The earliest was disclosed in 2022 and the latest in 2024. Each incident links to its company statement, filing, court document or post-mortem. | Disclosed | Organization | Access path | What was exposed | | --- | --- | --- | --- | | 2024-02-01 | [Cloudflare](/incidents/cloudflare-thanksgiving-2023/) | Leaked access keys and secrets | The company stated the threat actor accessed some documentation and a limited amount of source code on its self-hosted Atlassian server, and that the only production systems reachable with the stolen credentials were its Atlassian environment. | | 2023-10-23 | [1Password](/incidents/1password-okta-2023/) | Third-party access | The post says no 1Password user data was accessed and no compromise of user data or other sensitive systems was found. | | 2023-10-20 | [Okta](/incidents/okta-support-system-2023/) | Stolen user credentials and session tokens | Okta stated that files associated with 134 customers were accessed in its customer support system; some were HAR files containing session tokens, and the threat actor used them to hijack the sessions of 5 customers. | | 2023-10-20 | [BeyondTrust](/incidents/beyondtrust-okta-2023/) | Third-party access | The post says there was no impact or exposure to BeyondTrust’s infrastructure or customers. | | 2023-08-29 | [Retool](/incidents/retool-2023/) | Stolen user credentials and session tokens | The company stated that 27 cloud customer accounts were accessed without authorization (account takeovers, with emails changed and passwords reset); it stated no on-prem or managed accounts were affected. | | 2022-08-09 | [Cloudflare](/incidents/cloudflare-phishing-2022/) | Stolen user credentials and session tokens | The company stated that no Cloudflare systems were compromised. | | 2022-03-22 | [Okta (contractor Sitel)](/incidents/okta-sitel-2022/) | Third-party access | The post-mortem says screenshots were taken from a Sitel support engineer's computer; it states a maximum potential impact of 366 customers (approximately 2.5%) and that the Okta service was not breached. | ## How access was gained, entry by entry - [Cloudflare](/incidents/cloudflare-thanksgiving-2023/): The company stated the actor used one access token and three service account credentials taken in the October 2023 Okta compromise that the company had failed to rotate. - [1Password](/incidents/1password-okta-2023/): The company stated it detected suspicious activity on its Okta instance used to manage employee-facing apps on September 29, and later confirmed it was a result of the provider’s Support System breach. - [Okta](/incidents/okta-support-system-2023/): Okta stated the access used a service account stored in the support system itself, whose username and password had been saved in an employee's personal Google account on an Okta-managed laptop. - [BeyondTrust](/incidents/beyondtrust-okta-2023/): The company stated it detected an attacker trying to access an in-house Okta administrator account using a valid session cookie stolen from the provider’s support system. - [Retool](/incidents/retool-2023/): The company stated an employee was targeted by an SMS-based phishing attack followed by a call, and that the attacker added their own device to the employee's Okta account and used synced MFA codes to reach the VPN and internal admin systems. - [Cloudflare](/incidents/cloudflare-phishing-2022/): The company stated that employees received text messages pointing to a look-alike page for its identity provider, which asked for a username, password and a one-time code; individual employees did enter credentials, but physical security keys required for its applications stopped the attacker. - [Okta (contractor Sitel)](/incidents/okta-sitel-2022/): The post-mortem says an attacker obtained remote access, using RDP, to a computer owned and managed by Okta's sub-processor Sitel and used by a customer support engineer; the attacker did not take over an Okta account. ## Related pages - [Salesforce: cloud breaches in the ledger](/clouds/salesforce/) - [Cloud breaches: other providers and provider not public](/clouds/other/) - [AWS: cloud breaches in the ledger](/clouds/aws/) ## Sources - Cloudflare: https://blog.cloudflare.com/thanksgiving-2023-security-incident/ (read 16 Nov 2026) - 1Password: https://blog.1password.com/okta-incident/ (read 16 Nov 2026) - Okta: https://sec.okta.com/articles/2023/11/unauthorized-access-oktas-support-case-management-system-root-cause/ (read 16 Nov 2026) - BeyondTrust: https://www.beyondtrust.com/blog/entry/okta-support-unit-breach (read 16 Nov 2026) - Retool: https://retool.com/blog/mfa-isnt-mfa (read 16 Nov 2026) - Cloudflare: https://blog.cloudflare.com/2022-07-sms-phishing-attacks/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/clouds/okta/ --- # Cloud breaches involving Salesforce: 5 in the ledger Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. 5 incidents in the ledger involve Salesforce. The most common access path among them is third-party access. The earliest was disclosed in 2020 and the latest in 2026. Each incident links to its company statement, filing, court document or post-mortem. | Disclosed | Organization | Access path | What was exposed | | --- | --- | --- | --- | | 2026-06-23 | [8x8, Inc.](/incidents/8x8-salesforce-klue-2026/) | Third-party access | Certain information from the company's Salesforce system: fragmented contract and opportunity information, sales team notes, and contact information of customers and prospective customers | | 2025-09-02 | [Cloudflare](/incidents/cloudflare-salesforce-drift-2025/) | Third-party access | Salesforce case objects (customer support tickets and associated data: contact information, subject lines and case correspondence, not attachments); the company found 104 Cloudflare API tokens in the data | | 2025-08-05 | [Google](/incidents/google-salesforce-instance-2025/) | Stolen user credentials and session tokens | The update says a corporate Salesforce instance used to store contact information and related notes for small and medium businesses was affected; data retrieved was confined to basic and largely publicly available business information such as business names and contact details. | | 2025-03-12 | [Salesforce customers (multiple organizations)](/incidents/salesforce-customers-vishing-2025/) | Stolen user credentials and session tokens | Data exfiltrated from customer Salesforce accounts through a connected app; counts not public | | 2020-10-16 | [Chan Brothers Travel Pte Ltd](/incidents/chan-brothers-travel-pdpc-2019/) | Public storage, databases and services | The PDPC decision states the personal data of an estimated 5,593 individuals (names, email addresses and mobile numbers submitted through enquiry and feedback forms) was indexed, cached and searchable through web search engines. | ## How access was gained, entry by entry - [8x8, Inc.](/incidents/8x8-salesforce-klue-2026/): The filing says an unauthorized third party threat actor exploited the Klue third-party application programming integration connected to the company's Salesforce customer relationship management system. - [Cloudflare](/incidents/cloudflare-salesforce-drift-2025/): The company stated the threat actor obtained OAuth credentials associated with the Salesloft Drift chat agent's Salesforce integration, and used a stolen credential used by the Salesloft integration to log in to its Salesforce tenant. - [Google](/incidents/google-salesforce-instance-2025/): Google stated that the instance was impacted by UNC6040 activity described in the same post, which describes voice phishing in which employees are tricked into authorizing a malicious connected app or sharing credentials. - [Salesforce customers (multiple organizations)](/incidents/salesforce-customers-vishing-2025/): Salesforce stated threat actors impersonated IT support by phone, luring customers' employees and third-party support workers to phishing pages to steal credentials and MFA tokens, or to add a malicious connected app. - [Chan Brothers Travel Pte Ltd](/incidents/chan-brothers-travel-pdpc-2019/): The decision says the company hosted its site on Salesforce Community Cloud with the access configuration set to Public. A March 2018 release added an automated search engine optimisation feature, and the forms inherited the public setting, so the data was indexed. The company said it received no notice of the release; the decision found the release notes were published. ## Related pages - [Cloud breaches: other providers and provider not public](/clouds/other/) - [AWS: cloud breaches in the ledger](/clouds/aws/) - [Amazon Web Services: cloud breaches in the ledger](/clouds/amazon-web-services/) ## Sources - 8x8, Inc.: https://www.sec.gov/Archives/edgar/data/1023731/000102373126000084/eght-20260617.htm (read 16 Nov 2026) - Cloudflare: https://blog.cloudflare.com/response-to-salesloft-drift-incident/ (read 16 Nov 2026) - Google: https://cloud.google.com/blog/topics/threat-intelligence/voice-phishing-data-extortion (read 16 Nov 2026) - Salesforce customers (multiple organizations): https://www.salesforce.com/blog/protect-against-social-engineering/ (read 16 Nov 2026) - Chan Brothers Travel Pte Ltd: https://www.pdpc.gov.sg/assets/c4b13045-2ac1-42e8-bd76-b17f43d1ad47 (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/clouds/salesforce/ --- # Cloud breaches by year of disclosure Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. The ledger holds 111 incidents disclosed from 2019 to 2026. Each year page lists that year's incidents with the cloud or provider, the access path used and the primary sources, so the shift between paths across years can be read from the entries themselves, counted on 16 Nov 2026. | Year | Incidents | Leaked access keys and secrets | Stolen user credentials and session tokens | Over-permissioned identities | Public storage, databases and services | Cross-tenant access | Third-party access | Exposed network services and request paths | Retained access after departure or change | | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | | [2026](/years/2026/) | 11 | 0 | 6 | 0 | 0 | 0 | 5 | 1 | 1 | | [2025](/years/2025/) | 15 | 5 | 9 | 3 | 0 | 0 | 4 | 1 | 1 | | [2024](/years/2024/) | 14 | 3 | 3 | 3 | 0 | 0 | 2 | 3 | 0 | | [2023](/years/2023/) | 18 | 6 | 7 | 3 | 3 | 2 | 3 | 0 | 1 | | [2022](/years/2022/) | 26 | 7 | 14 | 2 | 4 | 0 | 5 | 1 | 2 | | [2021](/years/2021/) | 12 | 4 | 5 | 1 | 1 | 1 | 2 | 0 | 1 | | [2020](/years/2020/) | 12 | 1 | 3 | 1 | 6 | 0 | 4 | 0 | 1 | | [2019](/years/2019/) | 3 | 1 | 0 | 0 | 1 | 0 | 0 | 2 | 0 | ## Related pages - [Disclosed cloud attack techniques, with sources](/techniques/) - [Guides](/guides/) - [Glossary](/glossary/) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/years/ --- # Cloud security breaches disclosed in 2026: 11 in the ledger Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. The ledger holds 11 cloud incidents disclosed in 2026. They involve 5 clouds or providers, and the access path used most often was third-party access. Each entry rests on a company statement, a regulator filing, a court document or a provider post-mortem. | Disclosed | Organization | Cloud | Access path | | --- | --- | --- | --- | | 2026-09-08 | [Veradigm Inc.](/incidents/veradigm-vendor-api-2026/) | not public | Third-party access | | 2026-08-06 | [IEH Corporation](/incidents/ieh-microsoft-365-2026/) | Microsoft 365 | Stolen user credentials and session tokens | | 2026-07-31 | [Amgen](/incidents/amgen-cloud-environments-2026/) | not public | Not public | | 2026-07-02 | [AdaptHealth Corp.](/incidents/adapthealth-cloud-apps-2026/) | not public | Third-party access | | 2026-07-02 | [AdaptHealth](/incidents/adapthealth-contractor-session-2026/) | not public | Third-party access | | 2026-06-23 | [8x8, Inc.](/incidents/8x8-salesforce-klue-2026/) | Salesforce | Third-party access | | 2026-06-15 | [iRhythm Holdings, Inc.](/incidents/irhythm-hosted-apps-2026/) | not public | Stolen user credentials and session tokens | | 2026-05-08 | [SailPoint, Inc.](/incidents/sailpoint-github-2026/) | GitHub | Third-party access | | 2026-05-01 | [A Maryland medical system (called Company A in the indictment release)](/incidents/maryland-medical-system-m365-2016/) | Microsoft 365 | Stolen user credentials and session tokens | | 2026-04-24 | [ADT Inc.](/incidents/adt-cloud-environments-2026/) | not public | Not public | | 2026-01-08 | [People Central Pte. Ltd.](/incidents/people-central-pdpc-2024/) | AWS | Exposed network services and request paths | ## Related pages - [Cloud security breaches disclosed in 2025](/years/2025/) - [Cloud security breaches disclosed in 2024](/years/2024/) - [Cloud security breaches disclosed in 2023](/years/2023/) ## Sources - Veradigm Inc.: https://www.sec.gov/Archives/edgar/data/1124804/000119312526385249/mdrx-20260908.htm (read 16 Nov 2026) - IEH Corporation: https://www.sec.gov/Archives/edgar/data/50292/000117494726000761/form8k-36187_iehc.htm (read 16 Nov 2026) - Amgen: https://www.sec.gov/Archives/edgar/data/318154/000031815426000119/amgn-20260729.htm (read 16 Nov 2026) - AdaptHealth Corp.: https://www.sec.gov/Archives/edgar/data/1725255/000110465926080297/ahco-20260627x8k.htm (read 16 Nov 2026) - AdaptHealth: https://www.sec.gov/Archives/edgar/data/1725255/000110465926080297/ahco-20260627x8k.htm (read 16 Nov 2026) - 8x8, Inc.: https://www.sec.gov/Archives/edgar/data/1023731/000102373126000084/eght-20260617.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/years/2026/ --- # Cloud security breaches disclosed in 2025: 15 in the ledger Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. The ledger holds 15 cloud incidents disclosed in 2025. They involve 9 clouds or providers, and the access path used most often was stolen user credentials and session tokens. Each entry rests on a company statement, a regulator filing, a court document or a provider post-mortem. | Disclosed | Organization | Cloud | Access path | | --- | --- | --- | --- | | 2025-11-27 | [Mixpanel](/incidents/mixpanel-smishing-2025/) | Mixpanel | Stolen user credentials and session tokens | | 2025-11-06 | [Illuminate Education, Inc.](/incidents/illuminate-education-2021/) | not public (the FTC says a third-party cloud provider, not named in the press release) | Retained access after departure or change | | 2025-09-22 | [GitHub (npm registry)](/incidents/github-npm-shai-hulud-2025/) | GitHub | Stolen user credentials and session tokens | | 2025-09-02 | [Cloudflare](/incidents/cloudflare-salesforce-drift-2025/) | Salesforce | Third-party access | | 2025-08-26 | [Nx (Nrwl)](/incidents/nx-github-actions-2025/) | GitHub | Leaked access keys and secrets | | 2025-08-20 | [Salesloft (Drift)](/incidents/salesloft-drift-2025/) | AWS | Leaked access keys and secrets | | 2025-08-05 | [Google](/incidents/google-salesforce-instance-2025/) | Salesforce | Stolen user credentials and session tokens | | 2025-07-23 | [Amazon Web Services](/incidents/aws-amazon-q-extension-2025/) | AWS | Over-permissioned identities | | 2025-06-13 | [Zoomcar Holdings, Inc.](/incidents/zoomcar-cloud-2025/) | not public | Not public | | 2025-05-01 | [The Walt Disney Company](/incidents/disney-slack-2024/) | Slack | Stolen user credentials and session tokens | | 2025-04-28 | [Nova Scotia Power](/incidents/nova-scotia-power-opc-2025/) | not public | Stolen user credentials and session tokens | | 2025-03-27 | [Advanced Computer Software Group Ltd (health and care subsidiary)](/incidents/advanced-computer-software-2022/) | not public | Stolen user credentials and session tokens | | 2025-03-20 | [People Central Pte. Ltd. (Singapore, cloud-based HR software provider)](/incidents/people-central-aws-rdp-2024/) | Amazon Web Services | Exposed network services and request paths | | 2025-03-12 | [Salesforce customers (multiple organizations)](/incidents/salesforce-customers-vishing-2025/) | Salesforce | Stolen user credentials and session tokens | | 2025-01-07 | [PowerSchool](/incidents/powerschool-powersource-2024/) | PowerSchool | Stolen user credentials and session tokens | ## Related pages - [Cloud security breaches disclosed in 2024](/years/2024/) - [Cloud security breaches disclosed in 2023](/years/2023/) - [Cloud security breaches disclosed in 2022](/years/2022/) ## Sources - Mixpanel: https://mixpanel.com/blog/sms-security-incident/ (read 16 Nov 2026) - Illuminate Education, Inc.: https://web.archive.org/web/2024/https://www.ftc.gov/news-events/news/press-releases/2025/12/ftc-takes-action-against-education-technology-provider-failing-secure-students-personal-data (read 16 Nov 2026) - GitHub (npm registry): https://github.blog/security/supply-chain-security/our-plan-for-a-more-secure-npm-supply-chain/ (read 16 Nov 2026) - Cloudflare: https://blog.cloudflare.com/response-to-salesloft-drift-incident/ (read 16 Nov 2026) - Nx (Nrwl): https://nx.dev/blog/s1ngularity-postmortem (read 16 Nov 2026) - Salesloft (Drift): https://trust.salesloft.com/?uid=Drift+Security+Update (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/years/2025/ --- # Cloud security breaches disclosed in 2024: 14 in the ledger Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. The ledger holds 14 cloud incidents disclosed in 2024. They involve 8 clouds or providers, and the access path used most often was not public. Each entry rests on a company statement, a regulator filing, a court document or a provider post-mortem. | Disclosed | Organization | Cloud | Access path | | --- | --- | --- | --- | | 2024-12-08 | [BeyondTrust](/incidents/beyondtrust-remote-support-saas-2024/) | Amazon Web Services | Exposed network services and request paths | | 2024-10-22 | [Avaya Holdings Corp.](/incidents/avaya-cloud-file-sharing-2020/) | not public | Third-party access | | 2024-08-30 | [Verkada Inc.](/incidents/verkada-2021/) | Verkada (cloud-based security camera provider) | Over-permissioned identities | | 2024-07-12 | [AT&T](/incidents/att-third-party-cloud-2024/) | not public | Not public | | 2024-06-14 | [Advance Auto Parts](/incidents/advance-auto-parts-2024/) | not public | Not public | | 2024-06-13 | [Pure Storage](/incidents/pure-storage-snowflake-workspace-2024/) | Snowflake | Not public | | 2024-05-31 | [Live Nation Entertainment](/incidents/live-nation-ticketmaster-2024/) | not public | Not public | | 2024-05-31 | [Hugging Face](/incidents/huggingface-spaces-secrets-2024/) | Hugging Face | Not public | | 2024-05-30 | [Snowflake customer accounts (multiple organizations)](/incidents/snowflake-customer-accounts-2024/) | Snowflake | Stolen user credentials and session tokens | | 2024-05-01 | [Dropbox](/incidents/dropbox-sign-2024/) | not public | Over-permissioned identities | | 2024-04-22 | [Payroll2U Pte. Ltd.](/incidents/payroll2u-pdpc-2022/) | AWS | Stolen user credentials and session tokens | | 2024-02-01 | [Cloudflare](/incidents/cloudflare-thanksgiving-2023/) | Okta | Leaked access keys and secrets | | 2024-01-24 | [Hewlett Packard Enterprise](/incidents/hpe-midnight-blizzard-email-2024/) | not public | Not public | | 2024-01-19 | [Microsoft](/incidents/microsoft-midnight-blizzard-2024/) | Microsoft 365 | Stolen user credentials and session tokens | ## Related pages - [Cloud security breaches disclosed in 2023](/years/2023/) - [Cloud security breaches disclosed in 2022](/years/2022/) - [Cloud security breaches disclosed in 2021](/years/2021/) ## Sources - BeyondTrust: https://www.beyondtrust.com/remote-support-saas-service-security-investigation (read 16 Nov 2026) - Avaya Holdings Corp.: https://www.sec.gov/newsroom/press-releases/2024-174 (read 16 Nov 2026) - Verkada Inc.: https://web.archive.org/web/2025/https://www.ftc.gov/news-events/news/press-releases/2024/08/ftc-takes-action-against-security-camera-firm-verkada-over-charges-it-failed-secure-videos-other (read 16 Nov 2026) - AT&T: https://www.sec.gov/Archives/edgar/data/732717/000073271724000046/t-20240506.htm (read 16 Nov 2026) - Advance Auto Parts: https://www.sec.gov/Archives/edgar/data/1158449/000115844924000162/aap-20240523.htm (read 16 Nov 2026) - Pure Storage: https://www.sec.gov/Archives/edgar/data/1474432/000162828024027955/pstg-20240505.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/years/2024/ --- # Cloud security breaches disclosed in 2023: 18 in the ledger Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. The ledger holds 18 cloud incidents disclosed in 2023. They involve 9 clouds or providers, and the access path used most often was stolen user credentials and session tokens. Each entry rests on a company statement, a regulator filing, a court document or a provider post-mortem. | Disclosed | Organization | Cloud | Access path | | --- | --- | --- | --- | | 2023-12-11 | [A bank headquartered in San Francisco (not named in the U.S. Attorney release)](/incidents/sf-bank-cloud-engineer-retained-laptop-2020/) | not public | Retained access after departure or change | | 2023-11-16 | [Global Tel*Link Corp.](/incidents/global-tel-link-ftc-2020/) | not public | Public storage, databases and services | | 2023-11-07 | [Sumo Logic](/incidents/sumo-logic-aws-credential-2023/) | AWS | Leaked access keys and secrets | | 2023-10-23 | [1Password](/incidents/1password-okta-2023/) | Okta | Third-party access | | 2023-10-20 | [Okta](/incidents/okta-support-system-2023/) | Okta | Stolen user credentials and session tokens | | 2023-10-20 | [BeyondTrust](/incidents/beyondtrust-okta-2023/) | Okta | Third-party access | | 2023-10-06 | [23andMe](/incidents/23andme-2023/) | not public | Stolen user credentials and session tokens | | 2023-09-18 | [Microsoft](/incidents/microsoft-ai-sas-token-2023/) | Microsoft Azure | Leaked access keys and secrets | | 2023-08-30 | [Sourcegraph](/incidents/sourcegraph-2023/) | not public | Leaked access keys and secrets | | 2023-08-29 | [Retool](/incidents/retool-2023/) | Okta | Stolen user credentials and session tokens | | 2023-08-16 | [E-Commerce Enablers Pte. Ltd.](/incidents/ecommerce-enablers-pdpc-2020/) | AWS | Leaked access keys and secrets | | 2023-07-12 | [JumpCloud](/incidents/jumpcloud-orchestration-2023/) | JumpCloud | Stolen user credentials and session tokens | | 2023-07-11 | [Microsoft](/incidents/microsoft-storm-0558-2023/) | Microsoft 365 | Cross-tenant access | | 2023-06-16 | [1Health.io Inc. (formerly Vitagene, Inc.)](/incidents/vitagene-1health-ftc-2019/) | AWS | Public storage, databases and services | | 2023-03-23 | [GitHub](/incidents/github-rsa-ssh-key-2023/) | GitHub | Leaked access keys and secrets | | 2023-03-10 | [Eatigo International Pte. Ltd.](/incidents/eatigo-pdpc-2020/) | not public | Public storage, databases and services | | 2023-01-13 | [Mailchimp (Intuit)](/incidents/mailchimp-2023/) | Mailchimp | Stolen user credentials and session tokens | | 2023-01-04 | [CircleCI](/incidents/circleci-2023/) | CircleCI | Stolen user credentials and session tokens | ## Related pages - [Cloud security breaches disclosed in 2022](/years/2022/) - [Cloud security breaches disclosed in 2021](/years/2021/) - [Cloud security breaches disclosed in 2020](/years/2020/) ## Sources - A bank headquartered in San Francisco (not named in the U.S. Attorney release): https://web.archive.org/web/2026/https://www.justice.gov/usao-ndca/pr/disgruntled-cloud-engineer-sentenced-two-years-prison-intentionally-damaging-his (read 16 Nov 2026) - Global Tel*Link Corp.: https://web.archive.org/web/2024/https://www.ftc.gov/news-events/news/press-releases/2023/11/ftc-takes-action-against-global-tellink-corp-failing-adequately-secure-data-notify-consumers-after (read 16 Nov 2026) - Sumo Logic: https://web.archive.org/web/20240101000000/https://www.sumologic.com/security-response-center/ (read 16 Nov 2026) - 1Password: https://blog.1password.com/okta-incident/ (read 16 Nov 2026) - Okta: https://sec.okta.com/articles/2023/11/unauthorized-access-oktas-support-case-management-system-root-cause/ (read 16 Nov 2026) - BeyondTrust: https://www.beyondtrust.com/blog/entry/okta-support-unit-breach (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/years/2023/ --- # Cloud security breaches disclosed in 2022: 26 in the ledger Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. The ledger holds 26 cloud incidents disclosed in 2022. They involve 13 clouds or providers, and the access path used most often was stolen user credentials and session tokens. Each entry rests on a company statement, a regulator filing, a court document or a provider post-mortem. | Disclosed | Organization | Cloud | Access path | | --- | --- | --- | --- | | 2022-12-31 | [Slack](/incidents/slack-github-tokens-2022/) | GitHub | Stolen user credentials and session tokens | | 2022-12-22 | [LastPass](/incidents/lastpass-cloud-storage-2022/) | not public | Leaked access keys and secrets | | 2022-12-19 | [RedMart Limited](/incidents/redmart-pdpc-2020/) | AWS and Alibaba Cloud | Leaked access keys and secrets | | 2022-12-06 | [Rackspace Technology, Inc.](/incidents/rackspace-hosted-exchange-2022/) | Rackspace Hosted Exchange | Not public | | 2022-11-18 | [Farrer Park Hospital Pte Ltd](/incidents/farrer-park-hospital-pdpc-2018/) | Microsoft 365 | Not public | | 2022-11-01 | [Dropbox](/incidents/dropbox-github-phishing-2022/) | GitHub | Stolen user credentials and session tokens | | 2022-10-31 | [Chegg](/incidents/chegg-ftc-2022/) | not public | Retained access after departure or change | | 2022-10-24 | [Drizly](/incidents/drizly-ftc-2022/) | AWS | Stolen user credentials and session tokens | | 2022-10-19 | [Microsoft (misconfigured storage endpoint)](/incidents/microsoft-misconfigured-endpoint-2022/) | not public | Public storage, databases and services | | 2022-09-16 | [Uber](/incidents/uber-2022/) | Google Workspace | Stolen user credentials and session tokens | | 2022-09-16 | [GitHub users (phishing campaign reported by GitHub)](/incidents/github-circleci-phishing-2022/) | GitHub | Stolen user credentials and session tokens | | 2022-09-15 | [MyRepublic Limited](/incidents/myrepublic-pdpc-2021/) | AWS | Leaked access keys and secrets | | 2022-08-12 | [Mailchimp (Intuit)](/incidents/mailchimp-crypto-social-engineering-2022/) | Mailchimp | Stolen user credentials and session tokens | | 2022-08-10 | [Cisco Systems](/incidents/cisco-google-account-vishing-2022/) | Box | Stolen user credentials and session tokens | | 2022-08-09 | [Cloudflare](/incidents/cloudflare-phishing-2022/) | Okta | Stolen user credentials and session tokens | | 2022-08-07 | [Twilio](/incidents/twilio-2022/) | Twilio | Stolen user credentials and session tokens | | 2022-07-14 | [Crawfort Pte. Ltd.](/incidents/crawfort-pdpc-2020/) | AWS | Public storage, databases and services | | 2022-07-14 | [Quoine Pte Ltd](/incidents/quoine-pdpc-2020/) | not public | Third-party access | | 2022-06-30 | [Wegmans](/incidents/wegmans-azure-containers-2021/) | Microsoft Azure | Public storage, databases and services | | 2022-06-30 | [Wegmans Food Markets, Inc.](/incidents/wegmans-azure-container-2018/) | Microsoft Azure | Public storage, databases and services | | 2022-05-19 | [MGM Resorts International](/incidents/mgm-resorts-opc-2019/) | not public | Stolen user credentials and session tokens | | 2022-04-21 | [GeniusU Pte. Ltd.](/incidents/geniusu-pdpc-2021/) | GitHub | Leaked access keys and secrets | | 2022-04-15 | [Heroku (Salesforce) and Travis CI OAuth tokens](/incidents/heroku-travisci-oauth-2022/) | GitHub | Third-party access | | 2022-03-22 | [Okta (contractor Sitel)](/incidents/okta-sitel-2022/) | Okta | Third-party access | | 2022-03-22 | [Microsoft](/incidents/microsoft-dev-0537-2022/) | not public | Stolen user credentials and session tokens | | 2022-02-23 | [Alfi, Inc.](/incidents/alfi-former-executives-cloud-lockout-2022/) | Amazon Web Services | Retained access after departure or change | ## Related pages - [Cloud security breaches disclosed in 2021](/years/2021/) - [Cloud security breaches disclosed in 2020](/years/2020/) - [Cloud security breaches disclosed in 2019](/years/2019/) ## Sources - Slack: https://slack.com/blog/news/slack-security-update (read 16 Nov 2026) - LastPass: https://web.archive.org/web/2023/https://blog.lastpass.com/2022/12/notice-of-recent-security-incident/ (read 16 Nov 2026) - RedMart Limited: https://www.pdpc.gov.sg/assets/438fd68d-bcd9-4d25-8f77-09268e80a4e4 (read 16 Nov 2026) - Rackspace Technology, Inc.: https://www.sec.gov/Archives/edgar/data/1810019/000119312522301429/d429186d8k.htm (read 16 Nov 2026) - Farrer Park Hospital Pte Ltd: https://www.pdpc.gov.sg/assets/685c6f15-4fd9-49be-88e4-b86975ce7974 (read 16 Nov 2026) - Dropbox: https://dropbox.tech/security/a-recent-phishing-campaign-targeting-dropbox (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/years/2022/ --- # Cloud security breaches disclosed in 2021: 12 in the ledger Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. The ledger holds 12 cloud incidents disclosed in 2021. They involve 6 clouds or providers, and the access path used most often was stolen user credentials and session tokens. Each entry rests on a company statement, a regulator filing, a court document or a provider post-mortem. | Disclosed | Organization | Cloud | Access path | | --- | --- | --- | --- | | 2021-12-01 | [A publicly traded wireless communications manufacturer (called Company-1 in the U.S. Attorney release)](/incidents/ubiquiti-aws-github-2020/) | AWS | Over-permissioned identities | | 2021-11-22 | [GoDaddy](/incidents/godaddy-managed-wordpress-2021/) | GoDaddy Managed WordPress hosting | Stolen user credentials and session tokens | | 2021-11-11 | [Commeasure Pte Ltd (operates the RedDoorz booking platform)](/incidents/commeasure-reddoorz-pdpc-2020/) | AWS | Leaked access keys and secrets | | 2021-10-14 | [Stylez Pte Ltd](/incidents/stylez-icompare-pdpc-2019/) | not public | Public storage, databases and services | | 2021-10-14 | [The National Kidney Foundation](/incidents/national-kidney-foundation-pdpc-2020/) | Microsoft 365 | Stolen user credentials and session tokens | | 2021-09-21 | [Sendtech Pte. Ltd.](/incidents/sendtech-pdpc-2021/) | AWS | Leaked access keys and secrets | | 2021-08-30 | [Cetera Entities (Cetera Advisor Networks LLC, Cetera Investment Services LLC, Cetera Financial Specialists LLC, Cetera Advisors LLC, Cetera Investment Advisers LLC)](/incidents/cetera-email-account-takeovers-2017/) | not public | Stolen user credentials and session tokens | | 2021-08-30 | [Cambridge Investment Research Inc. and Cambridge Investment Research Advisors Inc.](/incidents/cambridge-email-account-takeovers-2018/) | not public | Stolen user credentials and session tokens | | 2021-08-30 | [KMS Financial Services Inc.](/incidents/kms-financial-email-account-takeovers-2018/) | not public | Stolen user credentials and session tokens | | 2021-08-27 | [Microsoft (Azure Cosmos DB)](/incidents/azure-cosmosdb-chaosdb-2021/) | Microsoft Azure | Cross-tenant access | | 2021-04-15 | [Codecov](/incidents/codecov-2021/) | Codecov | Leaked access keys and secrets | | 2021-03-16 | [Mimecast](/incidents/mimecast-solarwinds-certificate-2021/) | Microsoft 365 | Third-party access | ## Related pages - [Cloud security breaches disclosed in 2020](/years/2020/) - [Cloud security breaches disclosed in 2019](/years/2019/) - [Cloud security breaches disclosed in 2026](/years/2026/) ## Sources - A publicly traded wireless communications manufacturer (called Company-1 in the U.S. Attorney release): https://web.archive.org/web/2024/https://www.justice.gov/usao-sdny/pr/former-employee-technology-company-charged-stealing-confidential-data-and-extorting (read 16 Nov 2026) - GoDaddy: https://www.sec.gov/Archives/edgar/data/1609711/000160971121000122/gddyblogpostnov222021.htm (read 16 Nov 2026) - Commeasure Pte Ltd (operates the RedDoorz booking platform): https://www.pdpc.gov.sg/assets/f78cffe6-5efe-4757-b141-2d273702b812 (read 16 Nov 2026) - Stylez Pte Ltd: https://www.pdpc.gov.sg/assets/37845a6b-efc4-426d-bc1f-58e4e7da59ff (read 16 Nov 2026) - The National Kidney Foundation: https://www.pdpc.gov.sg/assets/8238cce1-16fd-4cec-b9ab-ffd8e8445a7b (read 16 Nov 2026) - Sendtech Pte. Ltd.: https://www.pdpc.gov.sg/assets/51342b7d-edaf-484b-82d5-e0dec632f4fc (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/years/2021/ --- # Cloud security breaches disclosed in 2020: 12 in the ledger Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. The ledger holds 12 cloud incidents disclosed in 2020. They involve 7 clouds or providers, and the access path used most often was public storage, databases and services. Each entry rests on a company statement, a regulator filing, a court document or a provider post-mortem. | Disclosed | Organization | Cloud | Access path | | --- | --- | --- | --- | | 2020-12-16 | [SkyMed International, Inc.](/incidents/skymed-ftc-2020/) | not public | Public storage, databases and services | | 2020-12-15 | [Ascension Data & Analytics](/incidents/ascension-data-2019/) | not public | Public storage, databases and services | | 2020-12-14 | [SolarWinds](/incidents/solarwinds-2020/) | Microsoft 365 | Third-party access | | 2020-10-16 | [Chan Brothers Travel Pte Ltd](/incidents/chan-brothers-travel-pdpc-2019/) | Salesforce | Public storage, databases and services | | 2020-10-16 | [MRI Diagnostics Pte Ltd and Clarity Radiology Pte Ltd](/incidents/mri-diagnostics-clarity-radiology-pdpc-2018/) | not public | Public storage, databases and services | | 2020-09-10 | [Singapore Medical Association](/incidents/singapore-medical-association-pdpc-2020/) | Microsoft 365 | Stolen user credentials and session tokens | | 2020-08-26 | [Cisco Systems](/incidents/cisco-webex-aws-2018/) | AWS | Retained access after departure or change | | 2020-07-22 | [Twilio](/incidents/twilio-2020/) | AWS | Public storage, databases and services | | 2020-07-18 | [Twitter](/incidents/twitter-2020/) | Twitter | Stolen user credentials and session tokens | | 2020-07-16 | [Blackbaud Inc.](/incidents/blackbaud-2020/) | Blackbaud (software provider; the New York Attorney General calls it a cloud company) | Over-permissioned identities | | 2020-07-11 | [LiveAuctioneers](/incidents/liveauctioneers-2020/) | AWS | Third-party access | | 2020-01-22 | [Microsoft](/incidents/microsoft-support-database-2019/) | Microsoft Azure | Public storage, databases and services | ## Related pages - [Cloud security breaches disclosed in 2019](/years/2019/) - [Cloud security breaches disclosed in 2026](/years/2026/) - [Cloud security breaches disclosed in 2025](/years/2025/) ## Sources - SkyMed International, Inc.: https://web.archive.org/web/2024/https://www.ftc.gov/news-events/news/press-releases/2020/12/company-provides-travel-emergency-services-settles-ftc-allegations-it-failed-secure-sensitive (read 16 Nov 2026) - Ascension Data & Analytics: https://www.ftc.gov/system/files/documents/cases/1923126ascension_complaint.pdf (read 16 Nov 2026) - SolarWinds: https://www.sec.gov/Archives/edgar/data/1739942/000162828020017451/swi-20201214.htm (read 16 Nov 2026) - Chan Brothers Travel Pte Ltd: https://www.pdpc.gov.sg/assets/c4b13045-2ac1-42e8-bd76-b17f43d1ad47 (read 16 Nov 2026) - MRI Diagnostics Pte Ltd and Clarity Radiology Pte Ltd: https://www.pdpc.gov.sg/assets/567882ff-bcde-4d39-a654-eae013e5889b (read 16 Nov 2026) - Singapore Medical Association: https://www.pdpc.gov.sg/assets/6db1d4a3-1e2f-4271-a7b6-292bfed8bfa1 (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/years/2020/ --- # Cloud security breaches disclosed in 2019: 3 in the ledger Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. The ledger holds 3 cloud incidents disclosed in 2019. They involve 1 clouds or providers, and the access path used most often was leaked access keys and secrets. Each entry rests on a company statement, a regulator filing, a court document or a provider post-mortem. | Disclosed | Organization | Cloud | Access path | | --- | --- | --- | --- | | 2019-08-27 | [Imperva](/incidents/imperva-2019/) | AWS | Leaked access keys and secrets | | 2019-07-29 | [Capital One](/incidents/capital-one-2019/) | AWS | Exposed network services and request paths | | 2019-06-28 | [Online Buddies (Jack’d dating app)](/incidents/online-buddies-jackd-s3-2018/) | AWS | Public storage, databases and services | ## Related pages - [Cloud security breaches disclosed in 2026](/years/2026/) - [Cloud security breaches disclosed in 2025](/years/2025/) - [Cloud security breaches disclosed in 2024](/years/2024/) ## Sources - Imperva: https://web.archive.org/web/20191101/https://www.imperva.com/blog/ceoblog/ (read 16 Nov 2026) - Capital One: https://www.sec.gov/Archives/edgar/data/927628/000092762819000262/exhibit991-pressrelease72919.htm (read 16 Nov 2026) - Online Buddies (Jack’d dating app): https://ag.ny.gov/press-release/2019/attorney-general-james-announces-settlement-dating-app-failure-secure-private-and (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/years/2019/ --- # Cloud security statistics: counts from the ledger, and figures from others Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. The ledger holds 111 cloud incidents from primary sources, counted on 16 Nov 2026; the most common access path is stolen user credentials and session tokens, in 47 entries (42 percent). Figures from third parties follow separately, each with its publisher, year and link, because they measure something different. ## Counted from the ledger Method: each incident counts once for its primary access path and once more for a secondary path where its sources describe one; 11 entries have no path stated. 111 incidents, computed on 16 Nov 2026. | Access path | Incidents | Share of incidents | | --- | --- | --- | | Leaked access keys and secrets | 27 | 24% | | Stolen user credentials and session tokens | 47 | 42% | | Over-permissioned identities | 13 | 12% | | Public storage, databases and services | 15 | 14% | | Cross-tenant access | 3 | 3% | | Third-party access | 25 | 23% | | Exposed network services and request paths | 8 | 7% | | Retained access after departure or change | 7 | 6% | | Not public (sources do not say) | 11 | 10% | | Source kind | Entries citing it | | --- | --- | | company statement | 39 | | regulator filing | 60 | | court document | 6 | | provider post mortem | 14 | ## Quoted from third parties These figures are not computed here; each is quoted with its publisher, year and link, and measures something different from the ledger. | Figure | What it says | Publisher | Year | | --- | --- | --- | --- | | 83% | Google's Cloud Threat Horizons Report H1 2026 says threat actors exploited identity issues to gain initial access in 83% of the incidents involving major cloud and SaaS-hosted environments it analyzed from the second half of 2025. ([source](https://cloud.google.com/security/report/resources/cloud-threat-horizons-report-h1-2026)) | Google Cloud Office of the CISO | 2026 | | 17% | Google's report says 17% of the cases it analyzed involved voice-based social engineering (vishing) as the initial access vector, and 12% involved email phishing. ([source](https://cloud.google.com/security/report/resources/cloud-threat-horizons-report-h1-2026)) | Google Cloud Office of the CISO | 2026 | | 21% | Google's report says 21% of the cases involved compromised trusted relationships with third parties, and a further 21% involved stolen human and non-human identities. ([source](https://cloud.google.com/security/report/resources/cloud-threat-horizons-report-h1-2026)) | Google Cloud Office of the CISO | 2026 | | 21% | Google's report says initial access through misconfiguration fell from 29.4% of incidents in the first half of 2025 to 21% in the second half. ([source](https://cloud.google.com/security/report/resources/cloud-threat-horizons-report-h1-2026)) | Google Cloud Office of the CISO | 2026 | | 43% | Datadog's State of Cloud Security says 43% of AWS IAM users have active keys created in the past year, as do 41% of Google Cloud service accounts and 66% of Entra ID app registrations. ([source](https://www.datadoghq.com/state-of-cloud-security/)) | Datadog | 2026 | | 54% | Datadog's report says 54% of organizations in AWS exclusively use keyless authentication for their GitHub Actions workflows, compared with 22% in Google Cloud and 5% in Azure. ([source](https://www.datadoghq.com/state-of-cloud-security/)) | Datadog | 2026 | | 63% | Datadog's report says 63% of organizations exclusively use single sign-on through AWS Identity Center and federated access, while 37% still use long-lived credentials through IAM users accessing the console. ([source](https://www.datadoghq.com/state-of-cloud-security/)) | Datadog | 2026 | | 32% | Mandiant's M-Trends 2026 post says exploits were the most common initial infection vector for the sixth consecutive year, at 32% of intrusions, with voice phishing second at 11%. ([source](https://cloud.google.com/blog/topics/threat-intelligence/m-trends-2026)) | Mandiant (Google Cloud) | 2026 | | 14 days | Mandiant's M-Trends 2026 post says the global median dwell time rose to 14 days from 11 days. ([source](https://cloud.google.com/blog/topics/threat-intelligence/m-trends-2026)) | Mandiant (Google Cloud) | 2026 | | 6% | Mandiant's M-Trends 2026 post says email phishing dropped to 6% of intrusions in 2025 as adversaries moved to voice-based social engineering. ([source](https://cloud.google.com/blog/topics/threat-intelligence/m-trends-2026)) | Mandiant (Google Cloud) | 2026 | | USD 4.99M | IBM's Cost of a Data Breach Report 2026 page states the global average cost of a data breach was USD 4.99 million, a 12% increase over the prior year. ([source](https://www.ibm.com/reports/data-breach)) | IBM and Ponemon Institute | 2026 | | 44% | IBM's X-Force Threat Intelligence Index 2026 page states a 44% year-over-year increase in the exploitation of public facing software or system applications. ([source](https://www.ibm.com/reports/threat-intelligence)) | IBM X-Force | 2026 | | 87% | Microsoft's Digital Defense Report 2025 page says adversaries are increasingly attacking the cloud, with destructive campaigns up 87%. ([source](https://www.microsoft.com/en-us/security/security-insider/threat-landscape/microsoft-digital-defense-report-2025)) | Microsoft | 2025 | | 266% | CrowdStrike's 2026 Global Threat Report page states a 266% increase in cloud-conscious intrusions by state-nexus threat actors. ([source](https://www.crowdstrike.com/en-us/resources/reports/global-threat-report/)) | CrowdStrike | 2026 | ## Related pages - [Correct an entry in the Cloud Breach Ledger](/ledger-corrections/) - [30 questions about public cloud incidents](/faq/) - [The Cloud Breach Ledger: the data, downloadable](/datasets/) ## Sources - Google Cloud Office of the CISO 2026: https://cloud.google.com/security/report/resources/cloud-threat-horizons-report-h1-2026 (read 16 Nov 2026) - Google Cloud Office of the CISO 2026: https://cloud.google.com/security/report/resources/cloud-threat-horizons-report-h1-2026 (read 16 Nov 2026) - Google Cloud Office of the CISO 2026: https://cloud.google.com/security/report/resources/cloud-threat-horizons-report-h1-2026 (read 16 Nov 2026) - Google Cloud Office of the CISO 2026: https://cloud.google.com/security/report/resources/cloud-threat-horizons-report-h1-2026 (read 16 Nov 2026) - Datadog 2026: https://www.datadoghq.com/state-of-cloud-security/ (read 16 Nov 2026) - Datadog 2026: https://www.datadoghq.com/state-of-cloud-security/ (read 16 Nov 2026) - Datadog 2026: https://www.datadoghq.com/state-of-cloud-security/ (read 16 Nov 2026) - Mandiant (Google Cloud) 2026: https://cloud.google.com/blog/topics/threat-intelligence/m-trends-2026 (read 16 Nov 2026) - Mandiant (Google Cloud) 2026: https://cloud.google.com/blog/topics/threat-intelligence/m-trends-2026 (read 16 Nov 2026) - Mandiant (Google Cloud) 2026: https://cloud.google.com/blog/topics/threat-intelligence/m-trends-2026 (read 16 Nov 2026) - IBM and Ponemon Institute 2026: https://www.ibm.com/reports/data-breach (read 16 Nov 2026) - IBM X-Force 2026: https://www.ibm.com/reports/threat-intelligence (read 16 Nov 2026) - Microsoft 2025: https://www.microsoft.com/en-us/security/security-insider/threat-landscape/microsoft-digital-defense-report-2025 (read 16 Nov 2026) - CrowdStrike 2026: https://www.crowdstrike.com/en-us/resources/reports/global-threat-report/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/statistics/ --- # Disclosed research techniques against cloud access Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Research techniques against cloud access, taken from the publishing organizations' own write-ups and logged with the publication date, publisher, a two-sentence neutral summary and the access path each concerns. They are disclosed research, not incidents, and they are not counted in the ledger. Researchers are never named, only organizations. | Published | Technique | Publisher | Access path | | --- | --- | --- | --- | | 2026-11-03 | [ReplicaHeist: cross-tenant object replication used to copy storage data](/techniques/replicaheist/) | Act Security research team | Cross-tenant access | | 2026-02-03 | [Credentials found in public buckets used to reach administrator rights](/techniques/ai-assisted-admin-in-8-minutes/) | Sysdig Threat Research Team | Leaked access keys and secrets | | 2025-02-12 | [whoAMI: image name confusion that runs an attacker's machine image](/techniques/whoami-ami-name-confusion/) | Datadog Security Labs | Third-party access | | 2025-01-29 | [An unauthenticated analytics database reachable from the internet](/techniques/deepseek-clickhouse/) | Wiz Research | Public storage, databases and services | | 2024-08-09 | [Bucket Monopoly: predictable bucket names claimed by an attacker](/techniques/bucket-monopoly/) | Aqua Security | Cross-tenant access | | 2024-07-24 | [ConfusedFunction: privilege escalation through a default build service account](/techniques/confusedfunction/) | Tenable Research | Over-permissioned identities | | 2024-02-28 | [Lateral movement through cloud APIs and compute instances](/techniques/cloud-lateral-movement/) | Palo Alto Networks Unit 42 | Over-permissioned identities | | 2024-01-25 | [Malicious OAuth applications used to keep access after an account is lost](/techniques/midnight-blizzard-oauth-apps/) | Microsoft Threat Intelligence | Retained access after departure or change | | 2024-01-19 | [Leaked IAM user keys used for storage theft and container crypto mining](/techniques/ecs-crypto-mining-leaked-keys/) | Datadog Security Labs | Leaked access keys and secrets | | 2023-09-18 | [A storage-account shared access token published with open-source models](/techniques/microsoft-ai-sas-token/) | Wiz Research | Leaked access keys and secrets | | 2023-07-11 | [SCARLETEEL: notebook containers exploited to steal cloud credentials](/techniques/scarleteel-fargate/) | Sysdig Threat Research Team | Exposed network services and request paths | | 2023-04-19 | [BrokenSesame: container escape and registry write access in a managed database service](/techniques/brokensesame/) | Wiz Research | Cross-tenant access | | 2023-04-11 | [Shared key authorization abused to escalate from storage to a subscription](/techniques/azure-shared-key-listkeys/) | Orca Security research | Over-permissioned identities | | 2023-03-29 | [BingBang: multi-tenant apps that accept sign-ins from any tenant](/techniques/bingbang/) | Wiz Research | Cross-tenant access | | 2022-03-07 | [AutoWarp: another customer's tokens reachable from an automation service](/techniques/autowarp/) | Orca Security research | Cross-tenant access | | 2021-11-11 | [ChaosDB: Cosmos DB keys obtained from a notebook feature](/techniques/chaosdb/) | Wiz Research | Cross-tenant access | | 2021-09-14 | [OMIGOD: unauthenticated remote code execution in a VM management agent](/techniques/omigod/) | Wiz Research | Exposed network services and request paths | | 2021-09-09 | [Azurescape: cross-account container takeover in a container service](/techniques/azurescape/) | Palo Alto Networks Unit 42 | Cross-tenant access | | 2019-11-19 | [IMDSv2: session tokens added against request-forgery access to instance credentials](/techniques/imdsv2-ssrf/) | Amazon Web Services | Exposed network services and request paths | ## Related pages - [Guides](/guides/) - [Glossary](/glossary/) - [News](/news/) ## Sources - ReplicaHeist: cross-tenant object replication used to copy storage data: https://act.security/resources/replicaheist-inside-azures-cross-tenant-data-exfiltration-technique (read 16 Nov 2026) - Azurescape: cross-account container takeover in a container service: https://unit42.paloaltonetworks.com/azure-container-instances/ (read 16 Nov 2026) - OMIGOD: unauthenticated remote code execution in a VM management agent: https://www.wiz.io/blog/omigod-critical-vulnerabilities-in-omi-azure (read 16 Nov 2026) - ChaosDB: Cosmos DB keys obtained from a notebook feature: https://www.wiz.io/blog/chaosdb-explained-azures-cosmos-db-vulnerability-walkthrough (read 16 Nov 2026) - AutoWarp: another customer's tokens reachable from an automation service: https://orca.security/resources/blog/autowarp-microsoft-azure-automation-service-vulnerability/ (read 16 Nov 2026) - BingBang: multi-tenant apps that accept sign-ins from any tenant: https://www.wiz.io/blog/azure-active-directory-bing-misconfiguration (read 16 Nov 2026) - BrokenSesame: container escape and registry write access in a managed database service: https://www.wiz.io/blog/brokensesame-accidental-write-permissions-to-private-registry-allowed-potential-r (read 16 Nov 2026) - Shared key authorization abused to escalate from storage to a subscription: https://orca.security/resources/blog/azure-shared-key-authorization-exploitation/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/techniques/ --- # ReplicaHeist: cross-tenant object replication used to copy storage data: the technique as the Act Security research team published it Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Act Security research team published "ReplicaHeist: cross-tenant object replication used to copy storage data" on 3 Nov 2026. The ledger files it under cross-tenant access. It is a disclosed research technique, not an incident. The source is the publisher's own write-up. - **Published:** 3 Nov 2026 - **Publisher:** Act Security research team - **Access path:** [Cross-tenant access](/access-paths/cross-tenant/) - **Source:** [the publisher's write-up](https://act.security/resources/replicaheist-inside-azures-cross-tenant-data-exfiltration-technique?utm_source=cloudsecuritybreaches.com&utm_medium=referral&utm_campaign=technique) Act's research team describes Azure object replication, a feature that copies blobs between storage accounts, as a channel that can copy data to a storage account in another tenant. The post says the allowCrossTenantReplication property is the boundary, and that accounts created before 15 December 2023 default to a null value that Azure treats as true. ## Related pages - [Azurescape: cross-account container takeover in a containe](/techniques/azurescape/) - [OMIGOD: unauthenticated remote code execution in a VM mana](/techniques/omigod/) - [ChaosDB: Cosmos DB keys obtained from a notebook feature](/techniques/chaosdb/) ## Sources - ReplicaHeist: cross-tenant object replication used to copy storage data: https://act.security/resources/replicaheist-inside-azures-cross-tenant-data-exfiltration-technique (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/techniques/replicaheist/ --- # Azurescape: cross-account container takeover in a container service: the technique as Palo Alto Networks Unit 42 published it Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Palo Alto Networks Unit 42 published "Azurescape: cross-account container takeover in a container service" on 9 Sep 2021. The ledger files it under cross-tenant access. It is a disclosed research technique, not an incident. The source is the publisher's own write-up. - **Published:** 9 Sep 2021 - **Publisher:** Palo Alto Networks Unit 42 - **Access path:** [Cross-tenant access](/access-paths/cross-tenant/) - **Source:** [the publisher's write-up](https://unit42.paloaltonetworks.com/azure-container-instances/) Unit 42 reported a chain that let a container escape its sandbox in Azure Container Instances and reach other customers' containers on the same platform. The post says Microsoft patched the service after the report and that Unit 42 has no knowledge of exploitation in the wild. ## Related pages - [OMIGOD: unauthenticated remote code execution in a VM mana](/techniques/omigod/) - [ChaosDB: Cosmos DB keys obtained from a notebook feature](/techniques/chaosdb/) - [AutoWarp: another customer's tokens reachable from an auto](/techniques/autowarp/) ## Sources - Azurescape: cross-account container takeover in a container service: https://unit42.paloaltonetworks.com/azure-container-instances/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/techniques/azurescape/ --- # OMIGOD: unauthenticated remote code execution in a VM management agent: the technique as the Wiz Research published it Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Wiz Research published "OMIGOD: unauthenticated remote code execution in a VM management agent" on 14 Sep 2021. Wiz Research reported four vulnerabilities in the OMI agent that Azure deploys on many Linux virtual machines, including an unauthenticated remote code execution as root. The post says over 65 percent of sampled Azure customers were exposed. - **Published:** 14 Sep 2021 - **Publisher:** Wiz Research - **Access path:** [Exposed network services and request paths](/access-paths/exposed-network-service/) - **Source:** [the publisher's write-up](https://www.wiz.io/blog/omigod-critical-vulnerabilities-in-omi-azure) Wiz Research reported four vulnerabilities in the OMI agent that Azure deploys on many Linux virtual machines, including an unauthenticated remote code execution as root. The post says over 65 percent of sampled Azure customers were exposed. ## Related pages - [ChaosDB: Cosmos DB keys obtained from a notebook feature](/techniques/chaosdb/) - [AutoWarp: another customer's tokens reachable from an auto](/techniques/autowarp/) - [BingBang: multi-tenant apps that accept sign-ins from any ](/techniques/bingbang/) ## Sources - OMIGOD: unauthenticated remote code execution in a VM management agent: https://www.wiz.io/blog/omigod-critical-vulnerabilities-in-omi-azure (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/techniques/omigod/ --- # ChaosDB: Cosmos DB keys obtained from a notebook feature: the technique as the Wiz Research published it Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Wiz Research published "ChaosDB: Cosmos DB keys obtained from a notebook feature" on 11 Nov 2021. The ledger files it under cross-tenant access. It is a disclosed research technique, not an incident. The source is the publisher's own write-up. Read on 16 Nov 2026. - **Published:** 11 Nov 2021 - **Publisher:** Wiz Research - **Access path:** [Cross-tenant access](/access-paths/cross-tenant/) - **Source:** [the publisher's write-up](https://www.wiz.io/blog/chaosdb-explained-azures-cosmos-db-vulnerability-walkthrough) Wiz Research describes how a Jupyter Notebook feature in Azure Cosmos DB led to root privileges on a container and then to primary keys for other customers' databases. The post says the flaw was disclosed to Microsoft in August 2021 and gave access to databases of several thousand customers. ## Related pages - [AutoWarp: another customer's tokens reachable from an auto](/techniques/autowarp/) - [BingBang: multi-tenant apps that accept sign-ins from any ](/techniques/bingbang/) - [BrokenSesame: container escape and registry write access i](/techniques/brokensesame/) ## Sources - ChaosDB: Cosmos DB keys obtained from a notebook feature: https://www.wiz.io/blog/chaosdb-explained-azures-cosmos-db-vulnerability-walkthrough (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/techniques/chaosdb/ --- # AutoWarp: another customer's tokens reachable from an automation service: the technique as the Orca Security research published it Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Orca Security research published "AutoWarp: another customer's tokens reachable from an automation service" on 7 Mar 2022. The ledger files it under cross-tenant access. It is a disclosed research technique, not an incident. The source is the publisher's own write-up. - **Published:** 7 Mar 2022 - **Publisher:** Orca Security research - **Access path:** [Cross-tenant access](/access-paths/cross-tenant/) - **Source:** [the publisher's write-up](https://orca.security/resources/blog/autowarp-microsoft-azure-automation-service-vulnerability/) Orca's research reports a flaw in Azure Automation that let it reach an internal server managing other customers' sandboxes and obtain authentication tokens for their accounts. The post gives a timeline with discovery on 6 December 2021, a fix on 10 December 2021 and public release on 7 March 2022. ## Related pages - [BingBang: multi-tenant apps that accept sign-ins from any ](/techniques/bingbang/) - [BrokenSesame: container escape and registry write access i](/techniques/brokensesame/) - [Shared key authorization abused to escalate from storage t](/techniques/azure-shared-key-listkeys/) ## Sources - AutoWarp: another customer's tokens reachable from an automation service: https://orca.security/resources/blog/autowarp-microsoft-azure-automation-service-vulnerability/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/techniques/autowarp/ --- # BingBang: multi-tenant apps that accept sign-ins from any tenant: the technique as the Wiz Research published it Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Wiz Research published "BingBang: multi-tenant apps that accept sign-ins from any tenant" on 29 Mar 2023. The ledger files it under cross-tenant access. It is a disclosed research technique, not an incident. The source is the publisher's own write-up. Read on 16 Nov 2026. - **Published:** 29 Mar 2023 - **Publisher:** Wiz Research - **Access path:** [Cross-tenant access](/access-paths/cross-tenant/) - **Source:** [the publisher's write-up](https://www.wiz.io/blog/azure-active-directory-bing-misconfiguration) Wiz Research found that applications configured as multi-tenant in Azure Active Directory, without their own check of who signs in, could be used by accounts from any tenant. The post says about 25 percent of multi-tenant applications it scanned were vulnerable, including a Microsoft content management system behind Bing.com. ## Related pages - [BrokenSesame: container escape and registry write access i](/techniques/brokensesame/) - [Shared key authorization abused to escalate from storage t](/techniques/azure-shared-key-listkeys/) - [SCARLETEEL: notebook containers exploited to steal cloud c](/techniques/scarleteel-fargate/) ## Sources - BingBang: multi-tenant apps that accept sign-ins from any tenant: https://www.wiz.io/blog/azure-active-directory-bing-misconfiguration (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/techniques/bingbang/ --- # BrokenSesame: container escape and registry write access in a managed database service: the technique as the Wiz Research published it Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Wiz Research published "BrokenSesame: container escape and registry write access in a managed database service" on 19 Apr 2023. The ledger files it under cross-tenant access. It is a disclosed research technique, not an incident. The source is the publisher's own write-up. - **Published:** 19 Apr 2023 - **Publisher:** Wiz Research - **Access path:** [Cross-tenant access](/access-paths/cross-tenant/) - **Source:** [the publisher's write-up](https://www.wiz.io/blog/brokensesame-accidental-write-permissions-to-private-registry-allowed-potential-r) Wiz Research describes a container escape in two Alibaba Cloud PostgreSQL services, combined with registry credentials that also had write permission, which could reach other tenants' databases. The post says Alibaba Cloud confirmed the issues were fully mitigated and that no customer data was compromised. ## Related pages - [Shared key authorization abused to escalate from storage t](/techniques/azure-shared-key-listkeys/) - [SCARLETEEL: notebook containers exploited to steal cloud c](/techniques/scarleteel-fargate/) - [A storage-account shared access token published with open-](/techniques/microsoft-ai-sas-token/) ## Sources - BrokenSesame: container escape and registry write access in a managed database service: https://www.wiz.io/blog/brokensesame-accidental-write-permissions-to-private-registry-allowed-potential-r (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/techniques/brokensesame/ --- # Shared key authorization abused to escalate from storage to a subscription: the technique as the Orca Security research published it Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Orca Security research published "Shared key authorization abused to escalate from storage to a subscription" on 11 Apr 2023. The ledger files it under over-permissioned identities. It is a disclosed research technique, not an incident. The source is the publisher's own write-up. - **Published:** 11 Apr 2023 - **Publisher:** Orca Security research - **Access path:** [Over-permissioned identities](/access-paths/over-permissioned-identity/) - **Source:** [the publisher's write-up](https://orca.security/resources/blog/azure-shared-key-authorization-exploitation/) Orca's research describes using the Azure listKeys permission on a storage account to alter a function's code, steal tokens of higher-privileged identities and move through a subscription. The post says Microsoft considered this a by-design risk and not a vulnerability, and recommends disabling shared key authorization. ## Related pages - [SCARLETEEL: notebook containers exploited to steal cloud c](/techniques/scarleteel-fargate/) - [A storage-account shared access token published with open-](/techniques/microsoft-ai-sas-token/) - [IMDSv2: session tokens added against request-forgery acces](/techniques/imdsv2-ssrf/) ## Sources - Shared key authorization abused to escalate from storage to a subscription: https://orca.security/resources/blog/azure-shared-key-authorization-exploitation/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/techniques/azure-shared-key-listkeys/ --- # SCARLETEEL: notebook containers exploited to steal cloud credentials: the technique as the Sysdig Threat Research Team published it Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Sysdig Threat Research Team published "SCARLETEEL: notebook containers exploited to steal cloud credentials" on 11 Jul 2023. Sysdig's team reports a continuing operation that exploited JupyterLab notebook containers in Kubernetes clusters and tried to steal AWS credentials, including attempts against the metadata service and in AWS Fargate. The post says the operation adapted its tools to bypass newer defenses. - **Published:** 11 Jul 2023 - **Publisher:** Sysdig Threat Research Team - **Access path:** [Exposed network services and request paths](/access-paths/exposed-network-service/) - **Source:** [the publisher's write-up](https://sysdig.com/blog/scarleteel-2-0/) Sysdig's team reports a continuing operation that exploited JupyterLab notebook containers in Kubernetes clusters and tried to steal AWS credentials, including attempts against the metadata service and in AWS Fargate. The post says the operation adapted its tools to bypass newer defenses. ## Related pages - [A storage-account shared access token published with open-](/techniques/microsoft-ai-sas-token/) - [IMDSv2: session tokens added against request-forgery acces](/techniques/imdsv2-ssrf/) - [Leaked IAM user keys used for storage theft and container ](/techniques/ecs-crypto-mining-leaked-keys/) ## Sources - SCARLETEEL: notebook containers exploited to steal cloud credentials: https://sysdig.com/blog/scarleteel-2-0/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/techniques/scarleteel-fargate/ --- # A storage-account shared access token published with open-source models: the technique as the Wiz Research published it Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Wiz Research published "A storage-account shared access token published with open-source models" on 18 Sep 2023. Wiz Research reports that a shared access signature link in a Microsoft GitHub repository was configured to cover a whole storage account of 38 terabytes, with full control permissions. The post says the account held workstation backups, secrets and over 30,000 internal Teams messages. - **Published:** 18 Sep 2023 - **Publisher:** Wiz Research - **Access path:** [Leaked access keys and secrets](/access-paths/leaked-access-key/) - **Source:** [the publisher's write-up](https://www.wiz.io/blog/38-terabytes-of-private-data-accidentally-exposed-by-microsoft-ai-researchers) Wiz Research reports that a shared access signature link in a Microsoft GitHub repository was configured to cover a whole storage account of 38 terabytes, with full control permissions. The post says the account held workstation backups, secrets and over 30,000 internal Teams messages. ## Related pages - [IMDSv2: session tokens added against request-forgery acces](/techniques/imdsv2-ssrf/) - [Leaked IAM user keys used for storage theft and container ](/techniques/ecs-crypto-mining-leaked-keys/) - [Malicious OAuth applications used to keep access after an ](/techniques/midnight-blizzard-oauth-apps/) ## Sources - A storage-account shared access token published with open-source models: https://www.wiz.io/blog/38-terabytes-of-private-data-accidentally-exposed-by-microsoft-ai-researchers (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/techniques/microsoft-ai-sas-token/ --- # IMDSv2: session tokens added against request-forgery access to instance credentials: the technique as Amazon Web Services published it Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Amazon Web Services published "IMDSv2: session tokens added against request-forgery access to instance credentials" on 19 Nov 2019. AWS introduced version 2 of its EC2 Instance Metadata Service, which requires a session token obtained with an HTTP PUT request. AWS says the change adds protection against open firewalls, reverse proxies and server-side request forgery reaching instance credentials. - **Published:** 19 Nov 2019 - **Publisher:** Amazon Web Services - **Access path:** [Exposed network services and request paths](/access-paths/exposed-network-service/) - **Source:** [the publisher's write-up](https://aws.amazon.com/blogs/security/defense-in-depth-open-firewalls-reverse-proxies-ssrf-vulnerabilities-ec2-instance-metadata-service/) AWS introduced version 2 of its EC2 Instance Metadata Service, which requires a session token obtained with an HTTP PUT request. AWS says the change adds protection against open firewalls, reverse proxies and server-side request forgery reaching instance credentials. ## Related pages - [Leaked IAM user keys used for storage theft and container ](/techniques/ecs-crypto-mining-leaked-keys/) - [Malicious OAuth applications used to keep access after an ](/techniques/midnight-blizzard-oauth-apps/) - [Lateral movement through cloud APIs and compute instances](/techniques/cloud-lateral-movement/) ## Sources - IMDSv2: session tokens added against request-forgery access to instance credentials: https://aws.amazon.com/blogs/security/defense-in-depth-open-firewalls-reverse-proxies-ssrf-vulnerabilities-ec2-instance-metadata-service/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/techniques/imdsv2-ssrf/ --- # Leaked IAM user keys used for storage theft and container crypto mining: the technique as the Datadog Security Labs published it Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Datadog Security Labs published "Leaked IAM user keys used for storage theft and container crypto mining" on 19 Jan 2024. The ledger files it under leaked access keys and secrets. It is a disclosed research technique, not an incident. The source is the publisher's own write-up. - **Published:** 19 Jan 2024 - **Publisher:** Datadog Security Labs - **Access path:** [Leaked access keys and secrets](/access-paths/leaked-access-key/) - **Source:** [the publisher's write-up](https://securitylabs.datadoghq.com/articles/tales-from-the-cloud-trenches-ecs-crypto-mining/) Datadog Security Labs describes attacker activity in AWS accounts in December 2023 and January 2024, most of it caused by a leaked IAM user access key, including S3 exfiltration and crypto mining on ECS. The post lists the steps seen and detection opportunities. ## Related pages - [Malicious OAuth applications used to keep access after an ](/techniques/midnight-blizzard-oauth-apps/) - [Lateral movement through cloud APIs and compute instances](/techniques/cloud-lateral-movement/) - [ConfusedFunction: privilege escalation through a default b](/techniques/confusedfunction/) ## Sources - Leaked IAM user keys used for storage theft and container crypto mining: https://securitylabs.datadoghq.com/articles/tales-from-the-cloud-trenches-ecs-crypto-mining/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/techniques/ecs-crypto-mining-leaked-keys/ --- # Malicious OAuth applications used to keep access after an account is lost: the technique as Microsoft Threat Intelligence published it Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Microsoft Threat Intelligence published "Malicious OAuth applications used to keep access after an account is lost" on 25 Jan 2024. The ledger files it under retained access after departure or change. It is a disclosed research technique, not an incident. - **Published:** 25 Jan 2024 - **Publisher:** Microsoft Threat Intelligence - **Access path:** [Retained access after departure or change](/access-paths/retained-access/) - **Source:** [the publisher's write-up](https://www.microsoft.com/en-us/security/blog/2024/01/25/midnight-blizzard-guidance-for-responders-on-nation-state-attack/) Microsoft's guidance says a threat actor it calls Midnight Blizzard compromised accounts to create and grant high permissions to OAuth applications. Microsoft says this let the actor keep access even if it lost the initially compromised account, and gives hunting queries for responders. ## Related pages - [Lateral movement through cloud APIs and compute instances](/techniques/cloud-lateral-movement/) - [ConfusedFunction: privilege escalation through a default b](/techniques/confusedfunction/) - [Bucket Monopoly: predictable bucket names claimed by an at](/techniques/bucket-monopoly/) ## Sources - Malicious OAuth applications used to keep access after an account is lost: https://www.microsoft.com/en-us/security/blog/2024/01/25/midnight-blizzard-guidance-for-responders-on-nation-state-attack/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/techniques/midnight-blizzard-oauth-apps/ --- # Lateral movement through cloud APIs and compute instances: the technique as Palo Alto Networks Unit 42 published it Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Palo Alto Networks Unit 42 published "Lateral movement through cloud APIs and compute instances" on 28 Feb 2024. The ledger files it under over-permissioned identities. It is a disclosed research technique, not an incident. The source is the publisher's own write-up. - **Published:** 28 Feb 2024 - **Publisher:** Palo Alto Networks Unit 42 - **Access path:** [Over-permissioned identities](/access-paths/over-permissioned-identity/) - **Source:** [the publisher's write-up](https://unit42.paloaltonetworks.com/cloud-lateral-movement-techniques/) Unit 42 describes lateral movement techniques it observed in the wild across AWS, Google Cloud and Azure, working at both the host level and the cloud API level. The post says cloud APIs offer attackers further mechanisms and that misconfigurations can open more. ## Related pages - [ConfusedFunction: privilege escalation through a default b](/techniques/confusedfunction/) - [Bucket Monopoly: predictable bucket names claimed by an at](/techniques/bucket-monopoly/) - [An unauthenticated analytics database reachable from the i](/techniques/deepseek-clickhouse/) ## Sources - Lateral movement through cloud APIs and compute instances: https://unit42.paloaltonetworks.com/cloud-lateral-movement-techniques/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/techniques/cloud-lateral-movement/ --- # ConfusedFunction: privilege escalation through a default build service account: the technique as the Tenable Research published it Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Tenable Research published "ConfusedFunction: privilege escalation through a default build service account" on 24 Jul 2024. The ledger files it under over-permissioned identities. It is a disclosed research technique, not an incident. The source is the publisher's own write-up. Read on 16 Nov 2026. - **Published:** 24 Jul 2024 - **Publisher:** Tenable Research - **Access path:** [Over-permissioned identities](/access-paths/over-permissioned-identity/) - **Source:** [the publisher's write-up](https://www.tenable.com/blog/confusedfunction-a-privilege-escalation-vulnerability-impacting-gcp-cloud-functions) Tenable Research reports that creating or updating a Google Cloud Function attached a default Cloud Build service account with excessive permissions, so a user able to deploy a function could escalate to it. The post says Google remediated it in part, for Cloud Build accounts created after mid-June 2024. ## Related pages - [Bucket Monopoly: predictable bucket names claimed by an at](/techniques/bucket-monopoly/) - [An unauthenticated analytics database reachable from the i](/techniques/deepseek-clickhouse/) - [whoAMI: image name confusion that runs an attacker's machi](/techniques/whoami-ami-name-confusion/) ## Sources - ConfusedFunction: privilege escalation through a default build service account: https://www.tenable.com/blog/confusedfunction-a-privilege-escalation-vulnerability-impacting-gcp-cloud-functions (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/techniques/confusedfunction/ --- # Bucket Monopoly: predictable bucket names claimed by an attacker: the technique as Aqua Security published it Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Aqua Security published "Bucket Monopoly: predictable bucket names claimed by an attacker" on 9 Aug 2024. Aqua's researchers describe a shadow resource attack, in which a service creates a storage bucket with a predictable name that an attacker registered first. The post says it found vulnerabilities in six AWS services and reported them to AWS in February 2024. - **Published:** 9 Aug 2024 - **Publisher:** Aqua Security - **Access path:** [Cross-tenant access](/access-paths/cross-tenant/) - **Source:** [the publisher's write-up](https://www.aquasec.com/blog/bucket-monopoly-breaching-aws-accounts-through-shadow-resources/) Aqua's researchers describe a shadow resource attack, in which a service creates a storage bucket with a predictable name that an attacker registered first. The post says it found vulnerabilities in six AWS services and reported them to AWS in February 2024. ## Related pages - [An unauthenticated analytics database reachable from the i](/techniques/deepseek-clickhouse/) - [whoAMI: image name confusion that runs an attacker's machi](/techniques/whoami-ami-name-confusion/) - [Credentials found in public buckets used to reach administ](/techniques/ai-assisted-admin-in-8-minutes/) ## Sources - Bucket Monopoly: predictable bucket names claimed by an attacker: https://www.aquasec.com/blog/bucket-monopoly-breaching-aws-accounts-through-shadow-resources/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/techniques/bucket-monopoly/ --- # An unauthenticated analytics database reachable from the internet: the technique as the Wiz Research published it Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Wiz Research published "An unauthenticated analytics database reachable from the internet" on 29 Jan 2025. The ledger files it under public storage, databases and services. It is a disclosed research technique, not an incident. The source is the publisher's own write-up. - **Published:** 29 Jan 2025 - **Publisher:** Wiz Research - **Access path:** [Public storage, databases and services](/access-paths/public-resource/) - **Source:** [the publisher's write-up](https://www.wiz.io/blog/wiz-research-uncovers-exposed-deepseek-database-leak) Wiz Research reports an open ClickHouse database tied to DeepSeek on two internet-facing hosts and non-standard ports 8123 and 9000. The post says it exposed over a million lines of logs with chat history and secret keys, and that the issue was disclosed to the company. ## Related pages - [whoAMI: image name confusion that runs an attacker's machi](/techniques/whoami-ami-name-confusion/) - [Credentials found in public buckets used to reach administ](/techniques/ai-assisted-admin-in-8-minutes/) - [ReplicaHeist: cross-tenant object replication used to copy](/techniques/replicaheist/) ## Sources - An unauthenticated analytics database reachable from the internet: https://www.wiz.io/blog/wiz-research-uncovers-exposed-deepseek-database-leak (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/techniques/deepseek-clickhouse/ --- # whoAMI: image name confusion that runs an attacker's machine image: the technique as the Datadog Security Labs published it Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Datadog Security Labs published "whoAMI: image name confusion that runs an attacker's machine image" on 12 Feb 2025. The ledger files it under third-party access. It is a disclosed research technique, not an incident. The source is the publisher's own write-up. - **Published:** 12 Feb 2025 - **Publisher:** Datadog Security Labs - **Access path:** [Third-party access](/access-paths/third-party-access/) - **Source:** [the publisher's write-up](https://securitylabs.datadoghq.com/articles/whoami-a-cloud-image-name-confusion-attack/) Datadog Security Labs describes software that looks up an Amazon Machine Image by name without naming the owner, so an attacker who publishes an image with a matching name can get it launched. The post says AWS announced Allowed AMIs on 1 December 2024, an allow list of trusted AMI providers. ## Related pages - [Credentials found in public buckets used to reach administ](/techniques/ai-assisted-admin-in-8-minutes/) - [ReplicaHeist: cross-tenant object replication used to copy](/techniques/replicaheist/) - [Azurescape: cross-account container takeover in a containe](/techniques/azurescape/) ## Sources - whoAMI: image name confusion that runs an attacker's machine image: https://securitylabs.datadoghq.com/articles/whoami-a-cloud-image-name-confusion-attack/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/techniques/whoami-ami-name-confusion/ --- # Credentials found in public buckets used to reach administrator rights: the technique as the Sysdig Threat Research Team published it Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Sysdig Threat Research Team published "Credentials found in public buckets used to reach administrator rights" on 3 Feb 2026. The ledger files it under leaked access keys and secrets. It is a disclosed research technique, not an incident. The source is the publisher's own write-up. - **Published:** 3 Feb 2026 - **Publisher:** Sysdig Threat Research Team - **Access path:** [Leaked access keys and secrets](/access-paths/leaked-access-key/) - **Source:** [the publisher's write-up](https://sysdig.com/blog/ai-assisted-cloud-intrusion-achieves-admin-access-in-8-minutes/) Sysdig's team reports an AWS intrusion observed on 28 November 2025 that began with valid test credentials found in public S3 buckets. It says the actor reached administrative privileges in under 10 minutes through Lambda code injection and moved across 19 principals. ## Related pages - [ReplicaHeist: cross-tenant object replication used to copy](/techniques/replicaheist/) - [Azurescape: cross-account container takeover in a containe](/techniques/azurescape/) - [OMIGOD: unauthenticated remote code execution in a VM mana](/techniques/omigod/) ## Sources - Credentials found in public buckets used to reach administrator rights: https://sysdig.com/blog/ai-assisted-cloud-intrusion-achieves-admin-access-in-8-minutes/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/techniques/ai-assisted-admin-in-8-minutes/ --- # Correct an entry in the ledger Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Each of the 111 entries below has its own correction link, which opens a message to editor@cloudsecuritybreaches.com with the organization and the entry name in the subject. A correction is published when a company statement, a regulator filing, a court document or a provider post-mortem supports it, and the change is dated on the entry. | Entry | Disclosed | Correction | | --- | --- | --- | | [Veradigm Inc.](/incidents/veradigm-vendor-api-2026/) | 2026-09-08 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Veradigm%20Inc.%20(veradigm-vendor-api-2026)) | | [IEH Corporation](/incidents/ieh-microsoft-365-2026/) | 2026-08-06 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20IEH%20Corporation%20(ieh-microsoft-365-2026)) | | [Amgen](/incidents/amgen-cloud-environments-2026/) | 2026-07-31 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Amgen%20(amgen-cloud-environments-2026)) | | [AdaptHealth Corp.](/incidents/adapthealth-cloud-apps-2026/) | 2026-07-02 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20AdaptHealth%20Corp.%20(adapthealth-cloud-apps-2026)) | | [AdaptHealth](/incidents/adapthealth-contractor-session-2026/) | 2026-07-02 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20AdaptHealth%20(adapthealth-contractor-session-2026)) | | [8x8, Inc.](/incidents/8x8-salesforce-klue-2026/) | 2026-06-23 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%208x8%2C%20Inc.%20(8x8-salesforce-klue-2026)) | | [iRhythm Holdings, Inc.](/incidents/irhythm-hosted-apps-2026/) | 2026-06-15 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20iRhythm%20Holdings%2C%20Inc.%20(irhythm-hosted-apps-2026)) | | [SailPoint, Inc.](/incidents/sailpoint-github-2026/) | 2026-05-08 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20SailPoint%2C%20Inc.%20(sailpoint-github-2026)) | | [A Maryland medical system (called Company A in the indictment release)](/incidents/maryland-medical-system-m365-2016/) | 2026-05-01 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20A%20Maryland%20medical%20system%20(called%20Company%20A%20in%20the%20indictment%20release)%20(maryland-medical-system-m365-2016)) | | [ADT Inc.](/incidents/adt-cloud-environments-2026/) | 2026-04-24 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20ADT%20Inc.%20(adt-cloud-environments-2026)) | | [People Central Pte. Ltd.](/incidents/people-central-pdpc-2024/) | 2026-01-08 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20People%20Central%20Pte.%20Ltd.%20(people-central-pdpc-2024)) | | [Mixpanel](/incidents/mixpanel-smishing-2025/) | 2025-11-27 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Mixpanel%20(mixpanel-smishing-2025)) | | [Illuminate Education, Inc.](/incidents/illuminate-education-2021/) | 2025-11-06 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Illuminate%20Education%2C%20Inc.%20(illuminate-education-2021)) | | [GitHub (npm registry)](/incidents/github-npm-shai-hulud-2025/) | 2025-09-22 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20GitHub%20(npm%20registry)%20(github-npm-shai-hulud-2025)) | | [Cloudflare](/incidents/cloudflare-salesforce-drift-2025/) | 2025-09-02 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Cloudflare%20(cloudflare-salesforce-drift-2025)) | | [Nx (Nrwl)](/incidents/nx-github-actions-2025/) | 2025-08-26 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Nx%20(Nrwl)%20(nx-github-actions-2025)) | | [Salesloft (Drift)](/incidents/salesloft-drift-2025/) | 2025-08-20 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Salesloft%20(Drift)%20(salesloft-drift-2025)) | | [Google](/incidents/google-salesforce-instance-2025/) | 2025-08-05 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Google%20(google-salesforce-instance-2025)) | | [Amazon Web Services](/incidents/aws-amazon-q-extension-2025/) | 2025-07-23 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Amazon%20Web%20Services%20(aws-amazon-q-extension-2025)) | | [Zoomcar Holdings, Inc.](/incidents/zoomcar-cloud-2025/) | 2025-06-13 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Zoomcar%20Holdings%2C%20Inc.%20(zoomcar-cloud-2025)) | | [The Walt Disney Company](/incidents/disney-slack-2024/) | 2025-05-01 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20The%20Walt%20Disney%20Company%20(disney-slack-2024)) | | [Nova Scotia Power](/incidents/nova-scotia-power-opc-2025/) | 2025-04-28 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Nova%20Scotia%20Power%20(nova-scotia-power-opc-2025)) | | [Advanced Computer Software Group Ltd (health and care subsidiary)](/incidents/advanced-computer-software-2022/) | 2025-03-27 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Advanced%20Computer%20Software%20Group%20Ltd%20(health%20and%20care%20subsidiary)%20(advanced-computer-software-2022)) | | [People Central Pte. Ltd. (Singapore, cloud-based HR software provider)](/incidents/people-central-aws-rdp-2024/) | 2025-03-20 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20People%20Central%20Pte.%20Ltd.%20(Singapore%2C%20cloud-based%20HR%20software%20provider)%20(people-central-aws-rdp-2024)) | | [Salesforce customers (multiple organizations)](/incidents/salesforce-customers-vishing-2025/) | 2025-03-12 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Salesforce%20customers%20(multiple%20organizations)%20(salesforce-customers-vishing-2025)) | | [PowerSchool](/incidents/powerschool-powersource-2024/) | 2025-01-07 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20PowerSchool%20(powerschool-powersource-2024)) | | [BeyondTrust](/incidents/beyondtrust-remote-support-saas-2024/) | 2024-12-08 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20BeyondTrust%20(beyondtrust-remote-support-saas-2024)) | | [Avaya Holdings Corp.](/incidents/avaya-cloud-file-sharing-2020/) | 2024-10-22 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Avaya%20Holdings%20Corp.%20(avaya-cloud-file-sharing-2020)) | | [Verkada Inc.](/incidents/verkada-2021/) | 2024-08-30 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Verkada%20Inc.%20(verkada-2021)) | | [AT&T](/incidents/att-third-party-cloud-2024/) | 2024-07-12 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20AT%26T%20(att-third-party-cloud-2024)) | | [Advance Auto Parts](/incidents/advance-auto-parts-2024/) | 2024-06-14 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Advance%20Auto%20Parts%20(advance-auto-parts-2024)) | | [Pure Storage](/incidents/pure-storage-snowflake-workspace-2024/) | 2024-06-13 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Pure%20Storage%20(pure-storage-snowflake-workspace-2024)) | | [Live Nation Entertainment](/incidents/live-nation-ticketmaster-2024/) | 2024-05-31 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Live%20Nation%20Entertainment%20(live-nation-ticketmaster-2024)) | | [Hugging Face](/incidents/huggingface-spaces-secrets-2024/) | 2024-05-31 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Hugging%20Face%20(huggingface-spaces-secrets-2024)) | | [Snowflake customer accounts (multiple organizations)](/incidents/snowflake-customer-accounts-2024/) | 2024-05-30 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Snowflake%20customer%20accounts%20(multiple%20organizations)%20(snowflake-customer-accounts-2024)) | | [Dropbox](/incidents/dropbox-sign-2024/) | 2024-05-01 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Dropbox%20(dropbox-sign-2024)) | | [Payroll2U Pte. Ltd.](/incidents/payroll2u-pdpc-2022/) | 2024-04-22 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Payroll2U%20Pte.%20Ltd.%20(payroll2u-pdpc-2022)) | | [Cloudflare](/incidents/cloudflare-thanksgiving-2023/) | 2024-02-01 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Cloudflare%20(cloudflare-thanksgiving-2023)) | | [Hewlett Packard Enterprise](/incidents/hpe-midnight-blizzard-email-2024/) | 2024-01-24 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Hewlett%20Packard%20Enterprise%20(hpe-midnight-blizzard-email-2024)) | | [Microsoft](/incidents/microsoft-midnight-blizzard-2024/) | 2024-01-19 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Microsoft%20(microsoft-midnight-blizzard-2024)) | | [A bank headquartered in San Francisco (not named in the U.S. Attorney release)](/incidents/sf-bank-cloud-engineer-retained-laptop-2020/) | 2023-12-11 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20A%20bank%20headquartered%20in%20San%20Francisco%20(not%20named%20in%20the%20U.S.%20Attorney%20release)%20(sf-bank-cloud-engineer-retained-laptop-2020)) | | [Global Tel*Link Corp.](/incidents/global-tel-link-ftc-2020/) | 2023-11-16 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Global%20Tel*Link%20Corp.%20(global-tel-link-ftc-2020)) | | [Sumo Logic](/incidents/sumo-logic-aws-credential-2023/) | 2023-11-07 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Sumo%20Logic%20(sumo-logic-aws-credential-2023)) | | [1Password](/incidents/1password-okta-2023/) | 2023-10-23 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%201Password%20(1password-okta-2023)) | | [Okta](/incidents/okta-support-system-2023/) | 2023-10-20 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Okta%20(okta-support-system-2023)) | | [BeyondTrust](/incidents/beyondtrust-okta-2023/) | 2023-10-20 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20BeyondTrust%20(beyondtrust-okta-2023)) | | [23andMe](/incidents/23andme-2023/) | 2023-10-06 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%2023andMe%20(23andme-2023)) | | [Microsoft](/incidents/microsoft-ai-sas-token-2023/) | 2023-09-18 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Microsoft%20(microsoft-ai-sas-token-2023)) | | [Sourcegraph](/incidents/sourcegraph-2023/) | 2023-08-30 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Sourcegraph%20(sourcegraph-2023)) | | [Retool](/incidents/retool-2023/) | 2023-08-29 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Retool%20(retool-2023)) | | [E-Commerce Enablers Pte. Ltd.](/incidents/ecommerce-enablers-pdpc-2020/) | 2023-08-16 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20E-Commerce%20Enablers%20Pte.%20Ltd.%20(ecommerce-enablers-pdpc-2020)) | | [JumpCloud](/incidents/jumpcloud-orchestration-2023/) | 2023-07-12 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20JumpCloud%20(jumpcloud-orchestration-2023)) | | [Microsoft](/incidents/microsoft-storm-0558-2023/) | 2023-07-11 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Microsoft%20(microsoft-storm-0558-2023)) | | [1Health.io Inc. (formerly Vitagene, Inc.)](/incidents/vitagene-1health-ftc-2019/) | 2023-06-16 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%201Health.io%20Inc.%20(formerly%20Vitagene%2C%20Inc.)%20(vitagene-1health-ftc-2019)) | | [GitHub](/incidents/github-rsa-ssh-key-2023/) | 2023-03-23 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20GitHub%20(github-rsa-ssh-key-2023)) | | [Eatigo International Pte. Ltd.](/incidents/eatigo-pdpc-2020/) | 2023-03-10 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Eatigo%20International%20Pte.%20Ltd.%20(eatigo-pdpc-2020)) | | [Mailchimp (Intuit)](/incidents/mailchimp-2023/) | 2023-01-13 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Mailchimp%20(Intuit)%20(mailchimp-2023)) | | [CircleCI](/incidents/circleci-2023/) | 2023-01-04 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20CircleCI%20(circleci-2023)) | | [Slack](/incidents/slack-github-tokens-2022/) | 2022-12-31 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Slack%20(slack-github-tokens-2022)) | | [LastPass](/incidents/lastpass-cloud-storage-2022/) | 2022-12-22 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20LastPass%20(lastpass-cloud-storage-2022)) | | [RedMart Limited](/incidents/redmart-pdpc-2020/) | 2022-12-19 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20RedMart%20Limited%20(redmart-pdpc-2020)) | | [Rackspace Technology, Inc.](/incidents/rackspace-hosted-exchange-2022/) | 2022-12-06 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Rackspace%20Technology%2C%20Inc.%20(rackspace-hosted-exchange-2022)) | | [Farrer Park Hospital Pte Ltd](/incidents/farrer-park-hospital-pdpc-2018/) | 2022-11-18 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Farrer%20Park%20Hospital%20Pte%20Ltd%20(farrer-park-hospital-pdpc-2018)) | | [Dropbox](/incidents/dropbox-github-phishing-2022/) | 2022-11-01 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Dropbox%20(dropbox-github-phishing-2022)) | | [Chegg](/incidents/chegg-ftc-2022/) | 2022-10-31 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Chegg%20(chegg-ftc-2022)) | | [Drizly](/incidents/drizly-ftc-2022/) | 2022-10-24 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Drizly%20(drizly-ftc-2022)) | | [Microsoft (misconfigured storage endpoint)](/incidents/microsoft-misconfigured-endpoint-2022/) | 2022-10-19 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Microsoft%20(misconfigured%20storage%20endpoint)%20(microsoft-misconfigured-endpoint-2022)) | | [Uber](/incidents/uber-2022/) | 2022-09-16 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Uber%20(uber-2022)) | | [GitHub users (phishing campaign reported by GitHub)](/incidents/github-circleci-phishing-2022/) | 2022-09-16 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20GitHub%20users%20(phishing%20campaign%20reported%20by%20GitHub)%20(github-circleci-phishing-2022)) | | [MyRepublic Limited](/incidents/myrepublic-pdpc-2021/) | 2022-09-15 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20MyRepublic%20Limited%20(myrepublic-pdpc-2021)) | | [Mailchimp (Intuit)](/incidents/mailchimp-crypto-social-engineering-2022/) | 2022-08-12 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Mailchimp%20(Intuit)%20(mailchimp-crypto-social-engineering-2022)) | | [Cisco Systems](/incidents/cisco-google-account-vishing-2022/) | 2022-08-10 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Cisco%20Systems%20(cisco-google-account-vishing-2022)) | | [Cloudflare](/incidents/cloudflare-phishing-2022/) | 2022-08-09 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Cloudflare%20(cloudflare-phishing-2022)) | | [Twilio](/incidents/twilio-2022/) | 2022-08-07 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Twilio%20(twilio-2022)) | | [Crawfort Pte. Ltd.](/incidents/crawfort-pdpc-2020/) | 2022-07-14 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Crawfort%20Pte.%20Ltd.%20(crawfort-pdpc-2020)) | | [Quoine Pte Ltd](/incidents/quoine-pdpc-2020/) | 2022-07-14 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Quoine%20Pte%20Ltd%20(quoine-pdpc-2020)) | | [Wegmans](/incidents/wegmans-azure-containers-2021/) | 2022-06-30 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Wegmans%20(wegmans-azure-containers-2021)) | | [Wegmans Food Markets, Inc.](/incidents/wegmans-azure-container-2018/) | 2022-06-30 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Wegmans%20Food%20Markets%2C%20Inc.%20(wegmans-azure-container-2018)) | | [MGM Resorts International](/incidents/mgm-resorts-opc-2019/) | 2022-05-19 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20MGM%20Resorts%20International%20(mgm-resorts-opc-2019)) | | [GeniusU Pte. Ltd.](/incidents/geniusu-pdpc-2021/) | 2022-04-21 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20GeniusU%20Pte.%20Ltd.%20(geniusu-pdpc-2021)) | | [Heroku (Salesforce) and Travis CI OAuth tokens](/incidents/heroku-travisci-oauth-2022/) | 2022-04-15 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Heroku%20(Salesforce)%20and%20Travis%20CI%20OAuth%20tokens%20(heroku-travisci-oauth-2022)) | | [Okta (contractor Sitel)](/incidents/okta-sitel-2022/) | 2022-03-22 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Okta%20(contractor%20Sitel)%20(okta-sitel-2022)) | | [Microsoft](/incidents/microsoft-dev-0537-2022/) | 2022-03-22 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Microsoft%20(microsoft-dev-0537-2022)) | | [Alfi, Inc.](/incidents/alfi-former-executives-cloud-lockout-2022/) | 2022-02-23 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Alfi%2C%20Inc.%20(alfi-former-executives-cloud-lockout-2022)) | | [A publicly traded wireless communications manufacturer (called Company-1 in the U.S. Attorney release)](/incidents/ubiquiti-aws-github-2020/) | 2021-12-01 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20A%20publicly%20traded%20wireless%20communications%20manufacturer%20(called%20Company-1%20in%20the%20U.S.%20Attorney%20release)%20(ubiquiti-aws-github-2020)) | | [GoDaddy](/incidents/godaddy-managed-wordpress-2021/) | 2021-11-22 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20GoDaddy%20(godaddy-managed-wordpress-2021)) | | [Commeasure Pte Ltd (operates the RedDoorz booking platform)](/incidents/commeasure-reddoorz-pdpc-2020/) | 2021-11-11 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Commeasure%20Pte%20Ltd%20(operates%20the%20RedDoorz%20booking%20platform)%20(commeasure-reddoorz-pdpc-2020)) | | [Stylez Pte Ltd](/incidents/stylez-icompare-pdpc-2019/) | 2021-10-14 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Stylez%20Pte%20Ltd%20(stylez-icompare-pdpc-2019)) | | [The National Kidney Foundation](/incidents/national-kidney-foundation-pdpc-2020/) | 2021-10-14 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20The%20National%20Kidney%20Foundation%20(national-kidney-foundation-pdpc-2020)) | | [Sendtech Pte. Ltd.](/incidents/sendtech-pdpc-2021/) | 2021-09-21 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Sendtech%20Pte.%20Ltd.%20(sendtech-pdpc-2021)) | | [Cetera Entities (Cetera Advisor Networks LLC, Cetera Investment Services LLC, Cetera Financial Specialists LLC, Cetera Advisors LLC, Cetera Investment Advisers LLC)](/incidents/cetera-email-account-takeovers-2017/) | 2021-08-30 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Cetera%20Entities%20(Cetera%20Advisor%20Networks%20LLC%2C%20Cetera%20Investment%20Services%20LLC%2C%20Cetera%20Financial%20Specialists%20LLC%2C%20Cetera%20Advisors%20LLC%2C%20Cetera%20Investment%20Advisers%20LLC)%20(cetera-email-account-takeovers-2017)) | | [Cambridge Investment Research Inc. and Cambridge Investment Research Advisors Inc.](/incidents/cambridge-email-account-takeovers-2018/) | 2021-08-30 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Cambridge%20Investment%20Research%20Inc.%20and%20Cambridge%20Investment%20Research%20Advisors%20Inc.%20(cambridge-email-account-takeovers-2018)) | | [KMS Financial Services Inc.](/incidents/kms-financial-email-account-takeovers-2018/) | 2021-08-30 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20KMS%20Financial%20Services%20Inc.%20(kms-financial-email-account-takeovers-2018)) | | [Microsoft (Azure Cosmos DB)](/incidents/azure-cosmosdb-chaosdb-2021/) | 2021-08-27 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Microsoft%20(Azure%20Cosmos%20DB)%20(azure-cosmosdb-chaosdb-2021)) | | [Codecov](/incidents/codecov-2021/) | 2021-04-15 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Codecov%20(codecov-2021)) | | [Mimecast](/incidents/mimecast-solarwinds-certificate-2021/) | 2021-03-16 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Mimecast%20(mimecast-solarwinds-certificate-2021)) | | [SkyMed International, Inc.](/incidents/skymed-ftc-2020/) | 2020-12-16 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20SkyMed%20International%2C%20Inc.%20(skymed-ftc-2020)) | | [Ascension Data & Analytics](/incidents/ascension-data-2019/) | 2020-12-15 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Ascension%20Data%20%26%20Analytics%20(ascension-data-2019)) | | [SolarWinds](/incidents/solarwinds-2020/) | 2020-12-14 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20SolarWinds%20(solarwinds-2020)) | | [Chan Brothers Travel Pte Ltd](/incidents/chan-brothers-travel-pdpc-2019/) | 2020-10-16 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Chan%20Brothers%20Travel%20Pte%20Ltd%20(chan-brothers-travel-pdpc-2019)) | | [MRI Diagnostics Pte Ltd and Clarity Radiology Pte Ltd](/incidents/mri-diagnostics-clarity-radiology-pdpc-2018/) | 2020-10-16 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20MRI%20Diagnostics%20Pte%20Ltd%20and%20Clarity%20Radiology%20Pte%20Ltd%20(mri-diagnostics-clarity-radiology-pdpc-2018)) | | [Singapore Medical Association](/incidents/singapore-medical-association-pdpc-2020/) | 2020-09-10 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Singapore%20Medical%20Association%20(singapore-medical-association-pdpc-2020)) | | [Cisco Systems](/incidents/cisco-webex-aws-2018/) | 2020-08-26 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Cisco%20Systems%20(cisco-webex-aws-2018)) | | [Twilio](/incidents/twilio-2020/) | 2020-07-22 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Twilio%20(twilio-2020)) | | [Twitter](/incidents/twitter-2020/) | 2020-07-18 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Twitter%20(twitter-2020)) | | [Blackbaud Inc.](/incidents/blackbaud-2020/) | 2020-07-16 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Blackbaud%20Inc.%20(blackbaud-2020)) | | [LiveAuctioneers](/incidents/liveauctioneers-2020/) | 2020-07-11 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20LiveAuctioneers%20(liveauctioneers-2020)) | | [Microsoft](/incidents/microsoft-support-database-2019/) | 2020-01-22 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Microsoft%20(microsoft-support-database-2019)) | | [Imperva](/incidents/imperva-2019/) | 2019-08-27 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Imperva%20(imperva-2019)) | | [Capital One](/incidents/capital-one-2019/) | 2019-07-29 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Capital%20One%20(capital-one-2019)) | | [Online Buddies (Jack’d dating app)](/incidents/online-buddies-jackd-s3-2018/) | 2019-06-28 | [Write about this entry](mailto:editor@cloudsecuritybreaches.com?subject=Correction%3A%20Online%20Buddies%20(Jack%E2%80%99d%20dating%20app)%20(online-buddies-jackd-s3-2018)) | ## Related pages - [30 questions about public cloud incidents](/faq/) - [The Cloud Breach Ledger: the data, downloadable](/datasets/) - [About](/about/) ## Sources - Veradigm Inc.: https://www.sec.gov/Archives/edgar/data/1124804/000119312526385249/mdrx-20260908.htm (read 16 Nov 2026) - IEH Corporation: https://www.sec.gov/Archives/edgar/data/50292/000117494726000761/form8k-36187_iehc.htm (read 16 Nov 2026) - Amgen: https://www.sec.gov/Archives/edgar/data/318154/000031815426000119/amgn-20260729.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/ledger-corrections/ --- # Ledger additions, edition by edition Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. The ledger is published in dated editions. The first edition is 16 November 2026 and covers incidents from 2019 to that date. Each edition fixes its sources and read dates, so a count can be cited as of that day. Later entries and corrections appear in the next edition with their dates. An edition is assembled from the ledger itself: every entry added since the previous edition, grouped by access path, each with its disclosure date and a link to the entry and its primary sources. Nothing appears in an edition that is not in the ledger, and an edition adds no commentary. There is no mailing list yet; the RSS feed carries each edition as it is published. - [Edition 1, 16 Nov 2026: the launch set of 111 entries](/editions/1/) ## Related pages - [Ledger edition 1: 111 incidents at launch](/editions/1/) - [Cloudflare reports cross-request data exposure in a Workers AI model](/news/cloudflare-workers-ai-kv-cache-exposure/) - [Cloudflare reports eight compromised intermediate certificate keys](/news/cloudflare-emudhra-certificates/) - [All of them](/news/) ## Sources - Veradigm Inc.: https://www.sec.gov/Archives/edgar/data/1124804/000119312526385249/mdrx-20260908.htm (read 16 Nov 2026) - IEH Corporation: https://www.sec.gov/Archives/edgar/data/50292/000117494726000761/form8k-36187_iehc.htm (read 16 Nov 2026) - Amgen: https://www.sec.gov/Archives/edgar/data/318154/000031815426000119/amgn-20260729.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/editions/ --- # Edition 1: the ledger at launch, 111 incidents Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Edition 1, dated 16 Nov 2026, is the launch set: 111 cloud incidents disclosed from 2019 to 2026, each from a company statement, regulator filing, court document or provider post-mortem. Later editions list only the entries added since the previous one, with their access paths. ## Leaked access keys and secrets: 27 - [GitHub (npm registry), 2025-09-22](/incidents/github-npm-shai-hulud-2025/) - [Cloudflare, 2025-09-02](/incidents/cloudflare-salesforce-drift-2025/) - [Nx (Nrwl), 2025-08-26](/incidents/nx-github-actions-2025/) - [Salesloft (Drift), 2025-08-20](/incidents/salesloft-drift-2025/) - [Amazon Web Services, 2025-07-23](/incidents/aws-amazon-q-extension-2025/) - [BeyondTrust, 2024-12-08](/incidents/beyondtrust-remote-support-saas-2024/) - [Dropbox, 2024-05-01](/incidents/dropbox-sign-2024/) - [Cloudflare, 2024-02-01](/incidents/cloudflare-thanksgiving-2023/) - [Sumo Logic, 2023-11-07](/incidents/sumo-logic-aws-credential-2023/) - [Microsoft, 2023-09-18](/incidents/microsoft-ai-sas-token-2023/) - [Sourcegraph, 2023-08-30](/incidents/sourcegraph-2023/) - [E-Commerce Enablers Pte. Ltd., 2023-08-16](/incidents/ecommerce-enablers-pdpc-2020/) - [Microsoft, 2023-07-11](/incidents/microsoft-storm-0558-2023/) - [GitHub, 2023-03-23](/incidents/github-rsa-ssh-key-2023/) - [LastPass, 2022-12-22](/incidents/lastpass-cloud-storage-2022/) - [RedMart Limited, 2022-12-19](/incidents/redmart-pdpc-2020/) - [Dropbox, 2022-11-01](/incidents/dropbox-github-phishing-2022/) - [Drizly, 2022-10-24](/incidents/drizly-ftc-2022/) - [MyRepublic Limited, 2022-09-15](/incidents/myrepublic-pdpc-2021/) - [GeniusU Pte. Ltd., 2022-04-21](/incidents/geniusu-pdpc-2021/) - [Heroku (Salesforce) and Travis CI OAuth tokens, 2022-04-15](/incidents/heroku-travisci-oauth-2022/) - [Commeasure Pte Ltd (operates the RedDoorz booking platform), 2021-11-11](/incidents/commeasure-reddoorz-pdpc-2020/) - [Sendtech Pte. Ltd., 2021-09-21](/incidents/sendtech-pdpc-2021/) - [Codecov, 2021-04-15](/incidents/codecov-2021/) - [Mimecast, 2021-03-16](/incidents/mimecast-solarwinds-certificate-2021/) - [LiveAuctioneers, 2020-07-11](/incidents/liveauctioneers-2020/) - [Imperva, 2019-08-27](/incidents/imperva-2019/) ## Stolen user credentials and session tokens: 47 - [Veradigm Inc., 2026-09-08](/incidents/veradigm-vendor-api-2026/) - [IEH Corporation, 2026-08-06](/incidents/ieh-microsoft-365-2026/) - [AdaptHealth Corp., 2026-07-02](/incidents/adapthealth-cloud-apps-2026/) - [AdaptHealth, 2026-07-02](/incidents/adapthealth-contractor-session-2026/) - [iRhythm Holdings, Inc., 2026-06-15](/incidents/irhythm-hosted-apps-2026/) - [A Maryland medical system (called Company A in the indictment release), 2026-05-01](/incidents/maryland-medical-system-m365-2016/) - [Mixpanel, 2025-11-27](/incidents/mixpanel-smishing-2025/) - [Illuminate Education, Inc., 2025-11-06](/incidents/illuminate-education-2021/) - [GitHub (npm registry), 2025-09-22](/incidents/github-npm-shai-hulud-2025/) - [Google, 2025-08-05](/incidents/google-salesforce-instance-2025/) - [The Walt Disney Company, 2025-05-01](/incidents/disney-slack-2024/) - [Nova Scotia Power, 2025-04-28](/incidents/nova-scotia-power-opc-2025/) - [Advanced Computer Software Group Ltd (health and care subsidiary), 2025-03-27](/incidents/advanced-computer-software-2022/) - [Salesforce customers (multiple organizations), 2025-03-12](/incidents/salesforce-customers-vishing-2025/) - [PowerSchool, 2025-01-07](/incidents/powerschool-powersource-2024/) - [Snowflake customer accounts (multiple organizations), 2024-05-30](/incidents/snowflake-customer-accounts-2024/) - [Payroll2U Pte. Ltd., 2024-04-22](/incidents/payroll2u-pdpc-2022/) - [Microsoft, 2024-01-19](/incidents/microsoft-midnight-blizzard-2024/) - [Okta, 2023-10-20](/incidents/okta-support-system-2023/) - [BeyondTrust, 2023-10-20](/incidents/beyondtrust-okta-2023/) - [23andMe, 2023-10-06](/incidents/23andme-2023/) - [Retool, 2023-08-29](/incidents/retool-2023/) - [JumpCloud, 2023-07-12](/incidents/jumpcloud-orchestration-2023/) - [Mailchimp (Intuit), 2023-01-13](/incidents/mailchimp-2023/) - [CircleCI, 2023-01-04](/incidents/circleci-2023/) - [Slack, 2022-12-31](/incidents/slack-github-tokens-2022/) - [RedMart Limited, 2022-12-19](/incidents/redmart-pdpc-2020/) - [Dropbox, 2022-11-01](/incidents/dropbox-github-phishing-2022/) - [Drizly, 2022-10-24](/incidents/drizly-ftc-2022/) - [Uber, 2022-09-16](/incidents/uber-2022/) - [GitHub users (phishing campaign reported by GitHub), 2022-09-16](/incidents/github-circleci-phishing-2022/) - [Mailchimp (Intuit), 2022-08-12](/incidents/mailchimp-crypto-social-engineering-2022/) - [Cisco Systems, 2022-08-10](/incidents/cisco-google-account-vishing-2022/) - [Cloudflare, 2022-08-09](/incidents/cloudflare-phishing-2022/) - [Twilio, 2022-08-07](/incidents/twilio-2022/) - [Quoine Pte Ltd, 2022-07-14](/incidents/quoine-pdpc-2020/) - [MGM Resorts International, 2022-05-19](/incidents/mgm-resorts-opc-2019/) - [GeniusU Pte. Ltd., 2022-04-21](/incidents/geniusu-pdpc-2021/) - [Microsoft, 2022-03-22](/incidents/microsoft-dev-0537-2022/) - [GoDaddy, 2021-11-22](/incidents/godaddy-managed-wordpress-2021/) - [The National Kidney Foundation, 2021-10-14](/incidents/national-kidney-foundation-pdpc-2020/) - [Cetera Entities (Cetera Advisor Networks LLC, Cetera Investment Services LLC, Cetera Financial Specialists LLC, Cetera Advisors LLC, Cetera Investment Advisers LLC), 2021-08-30](/incidents/cetera-email-account-takeovers-2017/) - [Cambridge Investment Research Inc. and Cambridge Investment Research Advisors Inc., 2021-08-30](/incidents/cambridge-email-account-takeovers-2018/) - [KMS Financial Services Inc., 2021-08-30](/incidents/kms-financial-email-account-takeovers-2018/) - [SolarWinds, 2020-12-14](/incidents/solarwinds-2020/) - [Singapore Medical Association, 2020-09-10](/incidents/singapore-medical-association-pdpc-2020/) - [Twitter, 2020-07-18](/incidents/twitter-2020/) ## Over-permissioned identities: 13 - [Nx (Nrwl), 2025-08-26](/incidents/nx-github-actions-2025/) - [Amazon Web Services, 2025-07-23](/incidents/aws-amazon-q-extension-2025/) - [Nova Scotia Power, 2025-04-28](/incidents/nova-scotia-power-opc-2025/) - [Verkada Inc., 2024-08-30](/incidents/verkada-2021/) - [Dropbox, 2024-05-01](/incidents/dropbox-sign-2024/) - [Microsoft, 2024-01-19](/incidents/microsoft-midnight-blizzard-2024/) - [Microsoft, 2023-09-18](/incidents/microsoft-ai-sas-token-2023/) - [E-Commerce Enablers Pte. Ltd., 2023-08-16](/incidents/ecommerce-enablers-pdpc-2020/) - [CircleCI, 2023-01-04](/incidents/circleci-2023/) - [Uber, 2022-09-16](/incidents/uber-2022/) - [Mailchimp (Intuit), 2022-08-12](/incidents/mailchimp-crypto-social-engineering-2022/) - [A publicly traded wireless communications manufacturer (called Company-1 in the U.S. Attorney release), 2021-12-01](/incidents/ubiquiti-aws-github-2020/) - [Blackbaud Inc., 2020-07-16](/incidents/blackbaud-2020/) ## Public storage, databases and services: 15 - [Global Tel*Link Corp., 2023-11-16](/incidents/global-tel-link-ftc-2020/) - [1Health.io Inc. (formerly Vitagene, Inc.), 2023-06-16](/incidents/vitagene-1health-ftc-2019/) - [Eatigo International Pte. Ltd., 2023-03-10](/incidents/eatigo-pdpc-2020/) - [Microsoft (misconfigured storage endpoint), 2022-10-19](/incidents/microsoft-misconfigured-endpoint-2022/) - [Crawfort Pte. Ltd., 2022-07-14](/incidents/crawfort-pdpc-2020/) - [Wegmans, 2022-06-30](/incidents/wegmans-azure-containers-2021/) - [Wegmans Food Markets, Inc., 2022-06-30](/incidents/wegmans-azure-container-2018/) - [Stylez Pte Ltd, 2021-10-14](/incidents/stylez-icompare-pdpc-2019/) - [SkyMed International, Inc., 2020-12-16](/incidents/skymed-ftc-2020/) - [Ascension Data & Analytics, 2020-12-15](/incidents/ascension-data-2019/) - [Chan Brothers Travel Pte Ltd, 2020-10-16](/incidents/chan-brothers-travel-pdpc-2019/) - [MRI Diagnostics Pte Ltd and Clarity Radiology Pte Ltd, 2020-10-16](/incidents/mri-diagnostics-clarity-radiology-pdpc-2018/) - [Twilio, 2020-07-22](/incidents/twilio-2020/) - [Microsoft, 2020-01-22](/incidents/microsoft-support-database-2019/) - [Online Buddies (Jack’d dating app), 2019-06-28](/incidents/online-buddies-jackd-s3-2018/) ## Cross-tenant access: 3 - [JumpCloud, 2023-07-12](/incidents/jumpcloud-orchestration-2023/) - [Microsoft, 2023-07-11](/incidents/microsoft-storm-0558-2023/) - [Microsoft (Azure Cosmos DB), 2021-08-27](/incidents/azure-cosmosdb-chaosdb-2021/) ## Third-party access: 25 - [Veradigm Inc., 2026-09-08](/incidents/veradigm-vendor-api-2026/) - [AdaptHealth Corp., 2026-07-02](/incidents/adapthealth-cloud-apps-2026/) - [AdaptHealth, 2026-07-02](/incidents/adapthealth-contractor-session-2026/) - [8x8, Inc., 2026-06-23](/incidents/8x8-salesforce-klue-2026/) - [SailPoint, Inc., 2026-05-08](/incidents/sailpoint-github-2026/) - [Cloudflare, 2025-09-02](/incidents/cloudflare-salesforce-drift-2025/) - [Salesloft (Drift), 2025-08-20](/incidents/salesloft-drift-2025/) - [Google, 2025-08-05](/incidents/google-salesforce-instance-2025/) - [Salesforce customers (multiple organizations), 2025-03-12](/incidents/salesforce-customers-vishing-2025/) - [Avaya Holdings Corp., 2024-10-22](/incidents/avaya-cloud-file-sharing-2020/) - [Cloudflare, 2024-02-01](/incidents/cloudflare-thanksgiving-2023/) - [Global Tel*Link Corp., 2023-11-16](/incidents/global-tel-link-ftc-2020/) - [1Password, 2023-10-23](/incidents/1password-okta-2023/) - [BeyondTrust, 2023-10-20](/incidents/beyondtrust-okta-2023/) - [Slack, 2022-12-31](/incidents/slack-github-tokens-2022/) - [Quoine Pte Ltd, 2022-07-14](/incidents/quoine-pdpc-2020/) - [MGM Resorts International, 2022-05-19](/incidents/mgm-resorts-opc-2019/) - [Heroku (Salesforce) and Travis CI OAuth tokens, 2022-04-15](/incidents/heroku-travisci-oauth-2022/) - [Okta (contractor Sitel), 2022-03-22](/incidents/okta-sitel-2022/) - [Codecov, 2021-04-15](/incidents/codecov-2021/) - [Mimecast, 2021-03-16](/incidents/mimecast-solarwinds-certificate-2021/) - [Ascension Data & Analytics, 2020-12-15](/incidents/ascension-data-2019/) - [SolarWinds, 2020-12-14](/incidents/solarwinds-2020/) - [MRI Diagnostics Pte Ltd and Clarity Radiology Pte Ltd, 2020-10-16](/incidents/mri-diagnostics-clarity-radiology-pdpc-2018/) - [LiveAuctioneers, 2020-07-11](/incidents/liveauctioneers-2020/) ## Exposed network services and request paths: 8 - [People Central Pte. Ltd., 2026-01-08](/incidents/people-central-pdpc-2024/) - [People Central Pte. Ltd. (Singapore, cloud-based HR software provider), 2025-03-20](/incidents/people-central-aws-rdp-2024/) - [BeyondTrust, 2024-12-08](/incidents/beyondtrust-remote-support-saas-2024/) - [Verkada Inc., 2024-08-30](/incidents/verkada-2021/) - [Payroll2U Pte. Ltd., 2024-04-22](/incidents/payroll2u-pdpc-2022/) - [Crawfort Pte. Ltd., 2022-07-14](/incidents/crawfort-pdpc-2020/) - [Imperva, 2019-08-27](/incidents/imperva-2019/) - [Capital One, 2019-07-29](/incidents/capital-one-2019/) ## Retained access after departure or change: 7 - [A Maryland medical system (called Company A in the indictment release), 2026-05-01](/incidents/maryland-medical-system-m365-2016/) - [Illuminate Education, Inc., 2025-11-06](/incidents/illuminate-education-2021/) - [A bank headquartered in San Francisco (not named in the U.S. Attorney release), 2023-12-11](/incidents/sf-bank-cloud-engineer-retained-laptop-2020/) - [Chegg, 2022-10-31](/incidents/chegg-ftc-2022/) - [Alfi, Inc., 2022-02-23](/incidents/alfi-former-executives-cloud-lockout-2022/) - [Sendtech Pte. Ltd., 2021-09-21](/incidents/sendtech-pdpc-2021/) - [Cisco Systems, 2020-08-26](/incidents/cisco-webex-aws-2018/) ## Related pages - [Cloudflare reports cross-request data exposure in a Workers AI model](/news/cloudflare-workers-ai-kv-cache-exposure/) - [Cloudflare reports eight compromised intermediate certificate keys](/news/cloudflare-emudhra-certificates/) - [Park Dental Partners confirms data acquisition and four lawsuits](/news/park-dental-partners-lawsuits/) - [All of them](/news/) ## Sources - Veradigm Inc.: https://www.sec.gov/Archives/edgar/data/1124804/000119312526385249/mdrx-20260908.htm (read 16 Nov 2026) - IEH Corporation: https://www.sec.gov/Archives/edgar/data/50292/000117494726000761/form8k-36187_iehc.htm (read 16 Nov 2026) - Amgen: https://www.sec.gov/Archives/edgar/data/318154/000031815426000119/amgn-20260729.htm (read 16 Nov 2026) - AdaptHealth Corp.: https://www.sec.gov/Archives/edgar/data/1725255/000110465926080297/ahco-20260627x8k.htm (read 16 Nov 2026) - AdaptHealth: https://www.sec.gov/Archives/edgar/data/1725255/000110465926080297/ahco-20260627x8k.htm (read 16 Nov 2026) - 8x8, Inc.: https://www.sec.gov/Archives/edgar/data/1023731/000102373126000084/eght-20260617.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/editions/1/ --- # Guides Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Guides to public cloud incidents built from company statements, regulator filings, court documents and provider post-mortems: the 2024 Snowflake customer incidents, Capital One, AWS-related incidents, cloud data breaches by year, recent additions and exposed-resource incidents. Each guide names its sources and the date they were read, 16 November 2026. - [Snowflake Data Breach: What the Filings Say](/snowflake-data-breach/): Court and company filings only; what is not public is marked. - [Capital One Breach: Company Statement and OCC Order](/capital-one-breach/): Company statement and regulator order only. - [AWS Data Breach: Incidents Involving AWS in the Ledger](/aws-data-breach/): Only incidents where a primary source names AWS. - [Cloud Data Breaches: What Was Exposed, As Stated](/cloud-data-breaches/): Counts quoted with their unit. - [Recent Cloud Security Breaches, 2025 and 2026](/recent-cloud-security-breaches/): Edition of 16 Nov 2026; newest first. - [Cloud Misconfiguration Breaches in the Ledger](/cloud-misconfiguration-breaches/): A list of incidents, not a definition or a tool review. ## Related pages - [Glossary](/glossary/) - [News](/news/) - [The eight access paths in cloud breaches](/access-paths/) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/guides/ --- # Snowflake data breach: what the company and customer filings say Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Snowflake's own SEC filings, read 16 Nov 2026, describe class actions filed from 13 June 2024 over customer data that threat actors accessed and exfiltrated, combined into one federal case in Montana. Customer filings from AT&T, Live Nation and Advance Auto Parts describe a third-party cloud environment without naming the provider. ## Snowflake's filings describe class actions over exfiltrated customer data Snowflake's quarterly report for the period ended July 31, 2024 says a class action was filed on June 13, 2024 in the United States District Court for the District of Montana. It alleges Snowflake failed to take reasonable measures to secure systems that contained consumer data, thereby allowing threat actors to access and exfiltrate customers' data (Form 10-Q, read 16 Nov 2026). The same report says numerous additional class actions making the same or similar allegations followed, against Snowflake or against customers whose consumer or employee data was exfiltrated. The claims include negligence, breach of fiduciary duty, breach of implied contract and unjust enrichment. The company wrote that it intends to defend against the claims. Snowflake's annual report for the year ended January 31, 2025 adds the later steps (Form 10-K, read 16 Nov 2026). On October 4, 2024 the Judicial Panel on Multidistrict Litigation combined the United States class actions into one case in the District of Montana. Plaintiffs filed a representative complaint on February 3, 2025. On March 11, 2025 they were granted until March 21, 2025 to file an amended complaint addressing pleading deficiencies the defendants identified. Two further class actions were pending in the Central District of California and the Supreme Court of British Columbia. ## Snowflake says customers choose and configure their own access controls Both filings carry a risk factor about customer security incidents. Snowflake wrote that its customers have experienced incidents in connection with use of its platform, including when those incidents are due to "vulnerabilities, policy violations, inadequate security controls, or credential exposures that we do not cause" (Form 10-K, read 16 Nov 2026). The 10-K says Snowflake is responsible for the security of its platform and underlying cloud infrastructure, while customers are responsible for selecting, enabling and configuring security controls for their own environments. It lists multi-factor authentication, network access policies and role-based access controls as tools the platform offers. It also says that where customers use third-party authentication tools, Snowflake has no visibility into whether controls such as MFA or network restrictions are enforced, and that customers who allow static access credentials are responsible for keeping them private and rotating them. That is the company's statement of its own position. The filings do not say which customers' accounts were accessed, which controls were absent on them, or how many accounts were involved. Those details are not public in the documents read. ## Three customer filings describe a third-party cloud environment and name no provider Three public companies reported incidents in spring 2024 involving a third-party cloud platform or database. None of the three filings names Snowflake or any other provider, and none says how access was gained, so this page does not attribute them to the provider. The ledger lists each with the access path third-party-access and the cloud as not public. | Company | Filing | What the filing says | Provider named | | --- | --- | --- | --- | | AT&T | Form 8-K, Item 1.05 (filed July 12, 2024) | Threat actors unlawfully accessed an AT&T workspace on a third-party cloud platform and, between April 14 and April 25, 2024, exfiltrated files with customer call and text records from about May 1 to October 31, 2022 and January 2, 2023. The Department of Justice approved delayed public reporting on May 9 and June 5, 2024. | No | | Live Nation Entertainment | Form 8-K, Item 8.01 (May 31, 2024) | The company identified unauthorized activity on May 20, 2024 within a third-party cloud database environment containing company data, primarily from its Ticketmaster subsidiary. A threat actor offered what it alleged to be user data for sale on May 27, 2024. | No | | Advance Auto Parts | Form 8-K, Item 8.01 (June 14, 2024) | The company identified unauthorized activity on May 23, 2024 within a third-party cloud database environment. A threat actor offered what it alleged to be company data for sale on June 4, 2024. Some files contain Social Security numbers or other government identification numbers of current and former applicants and employees. | No | ## Not public in the documents read Snowflake's own incident statement and customer guidance pages could not be retrieved on 16 Nov 2026 (the servers refused automated requests), so nothing on this page relies on them. No Department of Justice or court document on the incident was read; the class action descriptions above come from Snowflake's SEC reports. Where a figure such as the number of affected customers is absent from these filings, it is marked not public rather than estimated. ## Related pages - [Capital One Breach: Company Statement and OCC Order](/capital-one-breach/) - [AWS Data Breach: Incidents Involving AWS in the Ledger](/aws-data-breach/) - [Cloud Data Breaches: What Was Exposed, As Stated](/cloud-data-breaches/) - [All of them](/guides/) ## Sources - Snowflake Inc., Form 10-Q for the period ended July 31, 2024: https://www.sec.gov/Archives/edgar/data/1640147/000164014724000207/snow-20240731.htm (read 16 Nov 2026) - Snowflake Inc., Form 10-K for the year ended January 31, 2025: https://www.sec.gov/Archives/edgar/data/1640147/000164014725000052/snow-20250131.htm (read 16 Nov 2026) - AT&T Inc., Form 8-K, Item 1.05: https://www.sec.gov/Archives/edgar/data/732717/000073271724000046/t-20240506.htm (read 16 Nov 2026) - Live Nation Entertainment, Form 8-K, Item 8.01: https://www.sec.gov/Archives/edgar/data/1335258/000133525824000081/lyv-20240520.htm (read 16 Nov 2026) - Advance Auto Parts, Form 8-K, Item 8.01: https://www.sec.gov/Archives/edgar/data/1158449/000115844924000162/aap-20240523.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/snowflake-data-breach/ --- # Capital One breach: what the company and the OCC state Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Capital One stated that on 19 July 2019 it determined an outside individual had gained access, and that the incident affected about 100 million people in the United States and 6 million in Canada. The OCC consent order, signed 5 August 2020 and announced 6 August 2020, imposed an $80 million penalty over cloud risk management. Read 16 Nov 2026. ## Capital One stated an outside individual obtained data from March 2019 Capital One's incident page says that on July 19, 2019 the company determined an outside individual gained unauthorized access and obtained personal information about credit card customers and people who had applied for its credit card products. The page says the access occurred on March 22 and 23, 2019 (Capital One, read 16 Nov 2026). The company wrote that the incident affected approximately 100 million individuals in the United States and approximately 6 million in Canada. It also wrote that no credit card account numbers or log-in credentials were compromised, and that less than one percent of Social Security numbers were compromised. On discovery, the page says, the company fixed the issue and began working with federal law enforcement, and that the individual was captured by the FBI. It also reports the government's statement that it believes the data has been recovered and that there is no evidence it was used for fraud or shared. ## The company says a researcher reported a configuration vulnerability on July 17, 2019 In its FAQ the company says an external security researcher reported a "configuration vulnerability" through its program for security researchers to report vulnerabilities on July 17, 2019, and that its own investigation led to the July 19 determination. That is the only statement about the method of access in the company pages read. The page does not name the cloud provider, the service, the role or the request path involved. The method of access is therefore not public in the sources this ledger accepts, and this page does not describe it further. ## What Capital One said was exposed The company's page groups the data by category. The counts below are as the company stated them. | Data | As Capital One stated it | | --- | --- | | Credit card applications, 2005 through early 2019 | Names, addresses, zip or postal codes, phone numbers, email addresses, dates of birth and self-reported income; the largest category | | Customer status data | Credit scores, credit limits, balances, payment history and contact information | | Transaction data | Fragments from a total of 23 days during 2016, 2017 and 2018 | | U.S. Social Security numbers | About 140,000 of credit card customers; on January 27, 2021 the company found about 4,700 more among data accessed, not previously known | | Linked bank account numbers | About 80,000, of secured credit card customers | | Canadian Social Insurance numbers | Approximately 1 million | ## The OCC assessed an $80 million penalty on August 6, 2020 The Office of the Comptroller of the Currency announced the penalty in news release 2020-101 on August 6, 2020. The agency said it acted on the bank's failure to establish effective risk assessment processes before migrating significant information technology operations to the public cloud environment, and its failure to correct the deficiencies in a timely manner (OCC, read 16 Nov 2026). The consent order (docket AA-EC-20-49) records the Comptroller's findings, which the bank neither admitted nor denied. In or around 2015 the bank failed to establish effective risk assessment processes before the move, and failed to establish appropriate risk management for the cloud operating environment, including the design and implementation of certain network security controls, adequate data loss prevention controls and effective dispositioning of alerts. The order also finds that internal audit failed to identify numerous control weaknesses and gaps, and that the board failed to take effective action to hold management accountable. The order does not describe the 2019 incident itself in the pages read. It covers the bank's practices around the cloud move. ## Later steps and what remains not public The company's page records a class action settlement that a federal court preliminarily approved on February 7, 2022. A justice.gov page on the related criminal case returned no readable text on 16 Nov 2026, and no court record was read. For that reason the ledger does not assign this incident an access path from the documents above; the access path field reads not public until a company statement, filing or court document read by the ledger states the mechanism. ## Related pages - [AWS Data Breach: Incidents Involving AWS in the Ledger](/aws-data-breach/) - [Cloud Data Breaches: What Was Exposed, As Stated](/cloud-data-breaches/) - [Recent Cloud Security Breaches, 2025 and 2026](/recent-cloud-security-breaches/) - [All of them](/guides/) ## Sources - Capital One, Information on the Capital One cyber incident: https://www.capitalone.com/digital/facts2019/ (read 16 Nov 2026) - Capital One, 2019 cyber incident FAQ: https://www.capitalone.com/digital/facts2019/#faq (read 16 Nov 2026) - OCC news release 2020-101: https://www.occ.gov/news-issuances/news-releases/2020/nr-occ-2020-101.html (read 16 Nov 2026) - OCC consent order, AA-EC-20-49: https://www.occ.gov/static/enforcement-actions/ea2020-036.pdf (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/capital-one-breach/ --- # AWS data breach: incidents in the ledger that involve AWS Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Two ledger incidents name AWS in their primary source: the July 2025 Amazon Q Developer extension bulletin (AWS-2025-015) and the Salesloft Drift intrusion, whose summary says the actor reached Drift's AWS environment between March 22 and September 5, 2025. Capital One's sources describe a public cloud without naming AWS. Read 16 Nov 2026. ## AWS's own bulletin covers the Amazon Q Developer extension AWS security bulletin AWS-2025-015, first dated July 23, 2025, concerns the Amazon Q Developer for Visual Studio Code extension (CVE-2025-8217). AWS wrote that the extension had an inappropriately scoped GitHub token in its CodeBuild configuration, and that with that token a threat actor committed malicious code to the extension's open-source repository, where it was automatically included in a release (AWS, read 16 Nov 2026). AWS wrote that it revoked and replaced the credentials, removed the malicious code and released version 1.85.0. It also wrote that the code, distributed in version 1.84.0, was unsuccessful in executing because of a syntax error, so it made no changes to any services or customer environments. The bulletin adds that the code remains present in existing 1.84.0 installations, and tells customers to remove them and update. The bulletin describes a build-pipeline token, not a flaw in an AWS data service. The ledger classes it as over-permissioned-identity. ## Salesloft's investigation summary places an actor in Drift's AWS environment Salesloft's trust page lists an intrusion timeline running from March 22, 2025 to September 5, 2025. It says the actor used Salesloft GitHub personal access tokens for reconnaissance and secret enumeration, then accessed Drift's AWS environment and obtained OAuth tokens for Drift customers' technology integrations. It says the actor then used those tokens to access data through Drift integrations (Salesloft, read 16 Nov 2026). Among the remediation steps the page lists are an AWS lockdown covering security group and network ACL analysis, IAM role and policy review, container security, and logging and monitoring improvements. The page also says credentials such as OAuth tokens, employee passwords and application secrets were rotated, and that privileges were limited to the lowest levels necessary. The AWS environment here belonged to a customer of AWS; the page does not state that any AWS service failed. ## Where the sources name no provider Several ledger incidents involve a cloud environment that the primary sources do not name, so they cannot be listed as AWS incidents. Capital One's 2019 statement and the OCC's 2020 consent order refer to the cloud operating environment and the public cloud without naming the provider. AT&T, Live Nation and Advance Auto Parts describe a third-party cloud platform or database environment without naming it. This page lists none of them as AWS incidents because no source read states it. | Incident | Primary source | Provider named in source | Access path in the ledger | | --- | --- | --- | --- | | Amazon Q Developer extension, 2025 | AWS security bulletin | AWS | over-permissioned-identity | | Salesloft Drift, 2025 | Salesloft trust page | AWS | leaked-access-key | | Capital One, 2019 | Company page and OCC order | No | not public | | AT&T, Live Nation, Advance Auto Parts, 2024 | SEC Form 8-K filings | No | third-party-access | ## No AWS service breach found On 16 Nov 2026 the ledger holds no incident in which AWS stated that an outside party gained access to AWS's own infrastructure or to a service's customer data through a flaw in an AWS service. The AWS-related entries above involve a code-signing pipeline token and a customer's use of AWS. This is a statement about the sources read on that date, not a claim that none exists. ## Related pages - [Cloud Data Breaches: What Was Exposed, As Stated](/cloud-data-breaches/) - [Recent Cloud Security Breaches, 2025 and 2026](/recent-cloud-security-breaches/) - [Cloud Misconfiguration Breaches in the Ledger](/cloud-misconfiguration-breaches/) - [All of them](/guides/) ## Sources - AWS security bulletin AWS-2025-015: https://aws.amazon.com/security/security-bulletins/AWS-2025-015/ (read 16 Nov 2026) - Salesloft, Drift security update: https://trust.salesloft.com/?uid=Drift+Security+Update (read 16 Nov 2026) - Capital One, Information on the Capital One cyber incident: https://www.capitalone.com/digital/facts2019/ (read 16 Nov 2026) - OCC consent order, AA-EC-20-49: https://www.occ.gov/static/enforcement-actions/ea2020-036.pdf (read 16 Nov 2026) - AT&T Inc., Form 8-K, Item 1.05: https://www.sec.gov/Archives/edgar/data/732717/000073271724000046/t-20240506.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/aws-data-breach/ --- # Cloud data breaches: what was exposed, as the companies state it Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Company statements and filings read on 16 Nov 2026 give counts that mean different things: about 100 million people at Capital One, roughly 14,000 accounts at 23andMe, 134 customers at Okta, 133 accounts at Mailchimp, 27 accounts at Retool. This page lists each count with what it counts. ## The same word, breach, counts people, accounts, customers and tokens A count from one incident cannot be set beside a count from another without its unit. Capital One counted individuals. 23andMe counted accessed user accounts and the profiles visible through them. Okta counted customer organizations whose support files were viewed. Cloudflare counted API tokens found in exposed tickets. The table keeps each unit as the source wrote it and does not convert them. | Organization | Count as stated | Unit | Source type | | --- | --- | --- | --- | | Capital One (2019) | Approximately 100 million in the United States and approximately 6 million in Canada | Individuals affected | Company statement | | 23andMe (2023) | Roughly 14,000 user accounts (less than 0.1% of 14 million customers); about 5.5 million DNA Relatives profiles and about 1.4 million Family Tree profiles | Accounts accessed, and profiles seen through them | Company statement | | Okta (2023) | Files associated with 134 customers; session tokens used to hijack the sessions of 5 customers | Customer organizations | Company statement | | Mailchimp (2023) | 133 accounts | Accounts | Company statement | | Retool (2023) | 27 cloud customer accounts | Accounts taken over | Company statement | | Cloudflare (2025) | 104 Cloudflare API tokens found in the exposed case data | Tokens | Company statement | | Microsoft (2023) | Approximately 25 organizations in the public cloud, including government agencies | Organizations with email accessed | Provider post-mortem | ## Exposure is stated by data type, and some types are not public The 23andMe statement says the actor reached the 14,000 accounts using usernames and passwords that were also used on other websites, and that profile information in the connected relative features was visible through them (23andMe, read 16 Nov 2026). Mailchimp, the Intuit subsidiary, wrote that it found no evidence that Intuit systems or customer data beyond the 133 accounts were affected (Mailchimp, read 16 Nov 2026). Some filings give no data description at all. Advance Auto Parts' Form 8-K says only that some files contain personal information including Social Security numbers of current and former applicants and employees. AT&T's Form 8-K says the call and text records did not include content or Social Security numbers. Mixpanel's November 2025 statement lists no counts, and the ledger records the exposed field as not public. ## Some exposures were internal data, and the company said so Not every ledger entry involves customer records. Microsoft's September 2023 post on an overly permissive storage token says the exposed data included backups of two former employees' workstation profiles and internal Teams messages, and that no customer data was exposed (Microsoft, read 16 Nov 2026). Cloudflare's February 2024 post says the actor reached documentation and a limited amount of source code on its self-hosted Atlassian server, using an access token and three service account credentials it had not rotated after the October 2023 Okta compromise (Cloudflare, read 16 Nov 2026). These entries stay in the ledger because the access path, a leaked key or unrotated credential, matches the paths in customer-data incidents. ## What the counts leave out Counts are the companies' own, as of the date each statement was made, and several were revised. Capital One reported on February 22, 2021 that it found about 4,700 additional Social Security numbers among data already accessed. The ledger does not total the counts, because a total of individuals, accounts and tokens would not describe any single thing. ## Related pages - [Recent Cloud Security Breaches, 2025 and 2026](/recent-cloud-security-breaches/) - [Cloud Misconfiguration Breaches in the Ledger](/cloud-misconfiguration-breaches/) - [Snowflake Data Breach: What the Filings Say](/snowflake-data-breach/) - [All of them](/guides/) ## Sources - Capital One, Information on the Capital One cyber incident: https://www.capitalone.com/digital/facts2019/ (read 16 Nov 2026) - 23andMe, Addressing Data Security Concerns: https://blog.23andme.com/articles/addressing-data-security-concerns (read 16 Nov 2026) - Okta, root cause of unauthorized access to the support case management system: https://sec.okta.com/articles/2023/11/unauthorized-access-oktas-support-case-management-system-root-cause/ (read 16 Nov 2026) - Mailchimp, January 2023 security incident: https://mailchimp.com/january-2023-security-incident/ (read 16 Nov 2026) - Retool, MFA isn't MFA: https://retool.com/blog/mfa-isnt-mfa (read 16 Nov 2026) - Cloudflare, response to the Salesloft Drift incident: https://blog.cloudflare.com/response-to-salesloft-drift-incident/ (read 16 Nov 2026) - Cloudflare, Thanksgiving 2023 security incident: https://blog.cloudflare.com/thanksgiving-2023-security-incident/ (read 16 Nov 2026) - Microsoft MSRC, Storm-0558: https://www.microsoft.com/en-us/msrc/blog/2023/07/microsoft-mitigates-china-based-threat-actor-storm-0558-targeting-of-customer-email (read 16 Nov 2026) - Microsoft MSRC, overly permissive SAS token: https://www.microsoft.com/en-us/msrc/blog/2023/09/microsoft-mitigated-exposure-of-internal-information-in-a-storage-account-due-to-overly-permissive-sas-token (read 16 Nov 2026) - Mixpanel, SMS security incident: https://mixpanel.com/blog/sms-security-incident/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/cloud-data-breaches/ --- # Recent cloud security breaches in the ledger, 2025 to November 2026 Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Eight ledger entries were disclosed between March 2025 and August 2026, from the Salesforce vishing campaign to IEH Corporation's Form 8-K of 6 August 2026. Each is listed with its date, the source and the access path the source describes. Read 16 Nov 2026. ## Eight entries from 2025 and 2026, newest first Each row gives the date, the organization, what its primary source says about access, and the path the ledger assigns. The source is named in the second column and linked in the list below. | Disclosed | Organization | What the source says about access | Access path | | --- | --- | --- | --- | | Aug 6, 2026 | IEH Corporation (Form 8-K) | A phishing email carried a link disguised as a Microsoft document-sharing link; the user entered Microsoft 365 credentials into a fraudulent login page. One employee's mailbox was affected; the filing says no evidence currently exists of exfiltration. | stolen-user-credentials | | Jun 23, 2026 | 8x8, Inc. (Form 8-K) | An unauthorized third party exploited the Klue third-party integration connected to the company's Salesforce system on June 11 and 12, 2026. | third-party-access | | Nov 27, 2025 | Mixpanel (company post) | The company detected a smishing campaign on November 8, 2025 and revoked sessions and rotated credentials for impacted accounts. Data exposed: not public. | stolen-user-credentials | | Sep 2, 2025 | Cloudflare (company post) | The actor used OAuth credentials tied to the Salesloft Drift chat agent's Salesforce integration; 104 Cloudflare API tokens were found in the case data. | third-party-access | | Aug 26, 2025 | Nx (post-mortem) | A GitHub Actions injection vulnerability was used to steal an npm publishing token; malicious packages were live for 4 hours. | leaked-access-key | | Aug 20, 2025 | Salesloft (Drift) | The actor reached the Salesloft GitHub account, then Drift's AWS environment, and took OAuth tokens for customers' integrations. | leaked-access-key | | Jul 23, 2025 | Amazon Web Services (bulletin) | An inappropriately scoped GitHub token in a build configuration let the actor commit malicious code to an extension repository. | over-permissioned-identity | | Mar 12, 2025 | Salesforce customers (Salesforce post) | Callers impersonating IT support led employees to phishing pages or to add a malicious connected app; counts not public. | stolen-user-credentials | ## The two 2025 integration incidents share one root in Salesloft's report Cloudflare's post and Salesloft's investigation summary describe the same event from two sides. Salesloft says the actor used tokens taken from Drift's AWS environment; Cloudflare says it was among the customers whose Salesforce tenant was accessed with a stolen credential used by the Salesloft integration, and that the access ran from August 9 to 17, 2025 (Cloudflare and Salesloft, read 16 Nov 2026). The ledger lists both because each organization published its own statement, and it classes the supplier's entry by how the supplier was reached and the customer's by how the customer was reached. ## The 2026 entries are filings, which state less The two 2026 entries come from SEC Form 8-K filings. A filing records what the company chose to report: 8x8 reported fragmented contract and opportunity information, sales team notes and customer contact details, and said it and Klue, with support from Salesforce, disabled the integration. IEH reported that the affected mailbox held email messages, attachments, customer communications, purchase orders and engineering documentation. Neither filing gives a number of affected individuals; that field reads not public. ## Reading the list Among these eight entries, three are stolen-user-credentials, two are third-party-access, two are leaked-access-key and one is over-permissioned-identity. None is classed public-resource, cross-tenant, exposed-network-service or retained-access. That is a count of entries read for this edition, not a measure of how common each path is. New entries are added when a company statement, regulator filing, court document or provider post-mortem is found and read. ## Related pages - [Cloud Misconfiguration Breaches in the Ledger](/cloud-misconfiguration-breaches/) - [Snowflake Data Breach: What the Filings Say](/snowflake-data-breach/) - [Capital One Breach: Company Statement and OCC Order](/capital-one-breach/) - [All of them](/guides/) ## Sources - IEH Corporation, Form 8-K: https://www.sec.gov/Archives/edgar/data/50292/000117494726000761/form8k-36187_iehc.htm (read 16 Nov 2026) - 8x8, Inc., Form 8-K, Item 1.05: https://www.sec.gov/Archives/edgar/data/1023731/000102373126000084/eght-20260617.htm (read 16 Nov 2026) - Mixpanel, SMS security incident: https://mixpanel.com/blog/sms-security-incident/ (read 16 Nov 2026) - Cloudflare, response to the Salesloft Drift incident: https://blog.cloudflare.com/response-to-salesloft-drift-incident/ (read 16 Nov 2026) - Nx, s1ngularity post-mortem: https://nx.dev/blog/s1ngularity-postmortem (read 16 Nov 2026) - Salesloft, Drift security update: https://trust.salesloft.com/?uid=Drift+Security+Update (read 16 Nov 2026) - AWS security bulletin AWS-2025-015: https://aws.amazon.com/security/security-bulletins/AWS-2025-015/ (read 16 Nov 2026) - Salesforce, protect against social engineering: https://www.salesforce.com/blog/protect-against-social-engineering/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/recent-cloud-security-breaches/ --- # Cloud misconfiguration breaches: incidents where a source names a configuration or scoping fault Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Six ledger incidents have a primary source that names a configuration, scoping or permission fault: Microsoft's 2019 support database rules, a 2023 storage token, Amazon Q's build token, Dropbox Sign's service account, Microsoft's legacy test tenant and Capital One's configuration vulnerability. Read 16 Nov 2026. ## Six incidents where the source itself names the fault The ledger counts an incident here only when the company, regulator or provider says the cause involved a setting, a scope or a permission. It does not infer one. The table gives each source's wording and the date it was read. The list is of incidents. Tools that scan for such settings exist, and this page does not rank or review them. | Incident | Disclosed | What the source says | Closest path | | --- | --- | --- | --- | | Microsoft customer support database | Jan 22, 2020 | A change to the database's network security group on December 5, 2019 contained misconfigured security rules that enabled exposure; engineers fixed it on December 31, 2019 | public-resource | | Microsoft AI research storage account | Sep 18, 2023 | A URL in a public GitHub repository included an overly permissive Shared Access Signature token for an internal storage account | leaked-access-key | | Amazon Q Developer extension | Jul 23, 2025 | AWS found an inappropriately scoped GitHub token in the extension's CodeBuild configuration | over-permissioned-identity | | Dropbox Sign | May 1, 2024 | A compromised back-end service account held privileges across the production environment | over-permissioned-identity | | Microsoft corporate email (Midnight Blizzard) | Jan 19, 2024 | A password spray compromised a legacy non-production test tenant account whose permissions reached corporate email | stolen-user-credentials | | Capital One | Jul 2019 | The company calls the issue a configuration vulnerability reported by a researcher; the mechanism is not public in the sources read | not public | ## Microsoft's 2019 support database had misconfigured network rules Microsoft's January 2020 post says an investigation concluded that a change to the network security group of an internal customer support analytics database on December 5, 2019 contained misconfigured security rules. It says the investigation found no malicious use, that most customers did not have personal information exposed, and that the issue did not represent an exposure of its commercial cloud services (Microsoft, read 16 Nov 2026 via the Internet Archive). Microsoft wrote that it had solutions to help prevent this kind of mistake, but that they were not enabled for this database. It listed three actions: auditing network security rules for internal resources, expanding the mechanisms that detect rule misconfigurations, and adding alerting to service teams. ## A detection flagged Microsoft's 2023 storage token as a false positive Microsoft's September 2023 post says a Microsoft researcher inadvertently included a Shared Access Signature token in a blob store URL posted in a public GitHub repository while contributing to open-source AI models. Wiz reported it to Microsoft on June 22, 2023, and Microsoft revoked the token on June 24, 2023. The post says there was no security issue or vulnerability in Azure Storage or the SAS feature, and that no customer data was exposed. Microsoft also wrote that its own historical rescan of public repositories had detected the specific URL, but the finding was incorrectly marked as a false positive. It said the root cause was fixed (Microsoft, read 16 Nov 2026). ## Scope, not exposure, was the fault in three entries In AWS's bulletin the fault was a token scoped wider than its job. Dropbox's June 2024 post says the actor compromised a service account that was part of Sign's back-end and used it to reach the customer database. Microsoft's January 2024 post says the Midnight Blizzard actor used a legacy non-production test tenant account's permissions to reach corporate email. In all three the source describes access that was larger than needed, which the ledger records under over-permissioned-identity or the credential path. The OCC's 2020 consent order on Capital One lists network security controls, data loss prevention controls and alert dispositioning among its findings, as findings about the bank's cloud risk management rather than the cause of the 2019 incident. ## Related pages - [Snowflake Data Breach: What the Filings Say](/snowflake-data-breach/) - [Capital One Breach: Company Statement and OCC Order](/capital-one-breach/) - [AWS Data Breach: Incidents Involving AWS in the Ledger](/aws-data-breach/) - [All of them](/guides/) ## Sources - Microsoft MSRC, access misconfiguration for customer support database (Internet Archive copy): https://web.archive.org/web/2020/https://msrc-blog.microsoft.com/2020/01/22/access-misconfiguration-for-customer-support-database/ (read 16 Nov 2026) - Microsoft MSRC, overly permissive SAS token: https://www.microsoft.com/en-us/msrc/blog/2023/09/microsoft-mitigated-exposure-of-internal-information-in-a-storage-account-due-to-overly-permissive-sas-token (read 16 Nov 2026) - AWS security bulletin AWS-2025-015: https://aws.amazon.com/security/security-bulletins/AWS-2025-015/ (read 16 Nov 2026) - Dropbox Sign, a recent security incident: https://sign.dropbox.com/blog/a-recent-security-incident-involving-dropbox-sign (read 16 Nov 2026) - Microsoft MSRC, Midnight Blizzard: https://www.microsoft.com/en-us/msrc/blog/2024/01/microsoft-actions-following-attack-by-nation-state-actor-midnight-blizzard (read 16 Nov 2026) - Capital One, 2019 cyber incident FAQ: https://www.capitalone.com/digital/facts2019/#faq (read 16 Nov 2026) - OCC consent order, AA-EC-20-49: https://www.occ.gov/static/enforcement-actions/ea2020-036.pdf (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/cloud-misconfiguration-breaches/ --- # 30 terms, each with a sourced example Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Plain definitions of the terms used in the ledger, from access keys and session tokens to confused deputy, tenant and public access prevention. Each definition stands alone in one sentence and is tied to provider documentation or a primary filing that was read and dated, 16 November 2026. - [Access key](/glossary/access-key/): An access key is a long-term credential made of an access key ID and a secret access key that signs programmatic requests to a cloud provider's interfaces. - [Access path](/glossary/access-path/): An access path is the route an intruder used to reach cloud data, as the primary sources state it. - [Anonymous read access](/glossary/anonymous-read-access/): Anonymous read access lets anyone with a container's address download its contents without signing in. - [Block public access](/glossary/block-public-access/): Block public access is an account or bucket setting in Amazon S3 that overrides policies and permissions which would otherwise allow anonymous access to stored objects. - [Confused deputy](/glossary/confused-deputy/): The confused deputy problem is a flaw where a party without permission to act tricks a more privileged party into acting for it. - [Consent order](/glossary/consent-order/): A consent order is a regulator's order that a company agrees to without admitting or denying the findings. - [External ID](/glossary/external-id/): An external ID is an identifier placed in a role's trust policy so that a vendor assuming the role must present the value agreed for one customer. - [Form 8-K Item 1.05](/glossary/form-8-k-item-1-05/): Form 8-K Item 1.05 is the section of the U.S. - [Infostealer](/glossary/infostealer/): An infostealer is malware that collects saved passwords, browser cookies and session tokens from an infected device and sends them to its operator. - [Instance metadata service](/glossary/instance-metadata-service/): The instance metadata service is a local address that a virtual machine can query for facts about itself, including temporary credentials for the role attached to it. - [Least privilege](/glossary/least-privilege/): Least privilege is the practice of giving each identity only the actions on the resources that its task needs. - [Multi-factor authentication](/glossary/multi-factor-authentication/): Multi-factor authentication asks for a second proof of identity, such as a code or a hardware key, in addition to a password. - [Multitenant application](/glossary/multitenant-application/): A multitenant application is software registered to accept sign-ins from users in many organizations' tenants. - [Network security group](/glossary/network-security-group/): A network security group is Azure's rule set for allowing or denying inbound and outbound traffic to resources in a virtual network. - [OAuth token](/glossary/oauth-token/): An OAuth token is a credential that lets an application act for a user or a company in another service without holding a password. - [Offboarding](/glossary/offboarding/): Offboarding is the removal of a leaving person's or contractor's access to every account, key and connected application they held. - [OIDC trust](/glossary/oidc-trust/): An OIDC trust is a rule in a cloud account that accepts signed tokens from an outside identity provider and maps them to a role. - [Permissions boundary](/glossary/permissions-boundary/): A permissions boundary is a policy attached to an identity that sets the most that its other policies can ever grant. - [Provider post-mortem](/glossary/provider-post-mortem/): A provider post-mortem is a cloud or software-as-a-service provider's own written account of an incident on its platform, covering the timeline, cause and fix. - [Public access prevention](/glossary/public-access-prevention/): Public access prevention is a Google Cloud Storage setting that stops a bucket or a whole organization from granting access to anonymous users. - [Security group](/glossary/security-group/): A security group is a set of allow rules that controls which network traffic can reach a cloud resource and from where. - [Server-side request forgery](/glossary/server-side-request-forgery/): Server-side request forgery, or SSRF, is a flaw that makes an application send a request to an address the attacker picks. - [Service account key](/glossary/service-account-key/): A service account key is a file holding a private key that lets software authenticate as a Google Cloud service account from anywhere. - [Session token](/glossary/session-token/): A session token is the proof of a completed sign-in that a browser or application presents instead of a password. - [Shared access signature](/glossary/shared-access-signature/): A shared access signature is a signed link that gives a client delegated access to specific Azure Storage resources, with set permissions and a set expiry. - [Temporary credentials](/glossary/temporary-credentials/): Temporary credentials are short-lived keys and tokens, issued on request to a trusted identity or workload, that expire on their own after a set period. - [Tenant](/glossary/tenant/): A tenant is the boundary a provider draws around one customer's identities, applications and data inside a shared service. - [Third-party integration](/glossary/third-party-integration/): A third-party integration is an outside application connected to a company's cloud or software-as-a-service account by a token, key or role. - [Unused credential](/glossary/unused-credential/): An unused credential is a password, key or token that has not been used for a long period but still works. - [Workload identity federation](/glossary/workload-identity-federation/): Workload identity federation lets software running outside a cloud, such as a pipeline job, exchange a token from its own identity provider for short-lived cloud credentials. ## Related pages - [News](/news/) - [The eight access paths in cloud breaches](/access-paths/) - [Cloud breaches by cloud and provider](/clouds/) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/glossary/ --- # Access key Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. An access key is a long-term credential made of an access key ID and a secret access key that signs programmatic requests to a cloud provider's interfaces. Whoever holds both parts can act as the identity that owns the key until the key is deactivated or deleted. ## An example, sourced AWS's documentation says access keys are long-term credentials for an IAM user or the account root user, and that the secret part can be retrieved only when the key is created (read 16 Nov 2026). ## Why it matters here On Cloud Security Breaches the term appears in the rubric and on the platform pages; the example above was read on 16 Nov 2026 from the page under Sources. ## Related pages - [Temporary credentials: meaning and an example](/glossary/temporary-credentials/) - [Service account key: meaning and an example](/glossary/service-account-key/) - [Workload identity federation: meaning and an example](/glossary/workload-identity-federation/) - [All of them](/glossary/) ## Sources - https://docs.aws.amazon.com/IAM/latest/UserGuide/id_credentials_access-keys.html (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/glossary/access-key/ --- # Temporary credentials Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Temporary credentials are short-lived keys and tokens, issued on request to a trusted identity or workload, that expire on their own after a set period. A copy taken from a log or a repository stops working once the expiry time passes. ## An example, sourced AWS's documentation says its Security Token Service creates temporary security credentials for trusted users to control access to AWS resources (read 16 Nov 2026). ## Why it matters here On Cloud Security Breaches the term appears in the rubric and on the platform pages; the example above was read on 16 Nov 2026 from the page under Sources. ## Related pages - [Service account key: meaning and an example](/glossary/service-account-key/) - [Workload identity federation: meaning and an example](/glossary/workload-identity-federation/) - [OIDC trust: meaning and an example](/glossary/oidc-trust/) - [All of them](/glossary/) ## Sources - https://docs.aws.amazon.com/IAM/latest/UserGuide/id_credentials_temp.html (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/glossary/temporary-credentials/ --- # Service account key Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. A service account key is a file holding a private key that lets software authenticate as a Google Cloud service account from anywhere. Because it works outside the project that issued it, a copy in a repository or an image gives the finder the account's access. ## An example, sourced Google Cloud's documentation says service account keys can become a security risk if not managed carefully, and recommends a more secure alternative whenever possible (read 16 Nov 2026). ## Why it matters here On Cloud Security Breaches the term appears in the rubric and on the platform pages; the example above was read on 16 Nov 2026 from the page under Sources. ## Related pages - [Workload identity federation: meaning and an example](/glossary/workload-identity-federation/) - [OIDC trust: meaning and an example](/glossary/oidc-trust/) - [Session token: meaning and an example](/glossary/session-token/) - [All of them](/glossary/) ## Sources - https://cloud.google.com/iam/docs/best-practices-for-managing-service-account-keys (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/glossary/service-account-key/ --- # Workload identity federation Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Workload identity federation lets software running outside a cloud, such as a pipeline job, exchange a token from its own identity provider for short-lived cloud credentials. It removes the need to store a long-term key in the pipeline's secrets. The pipeline then holds nothing worth stealing between runs. ## An example, sourced Google Cloud's documentation describes Workload Identity Federation as a way to federate identities for external workloads (read 16 Nov 2026). ## Why it matters here On Cloud Security Breaches the term appears in the rubric and on the platform pages; the example above was read on 16 Nov 2026 from the page under Sources. ## Related pages - [OIDC trust: meaning and an example](/glossary/oidc-trust/) - [Session token: meaning and an example](/glossary/session-token/) - [OAuth token: meaning and an example](/glossary/oauth-token/) - [All of them](/glossary/) ## Sources - https://cloud.google.com/iam/docs/workload-identity-federation (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/glossary/workload-identity-federation/ --- # OIDC trust Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. An OIDC trust is a rule in a cloud account that accepts signed tokens from an outside identity provider and maps them to a role. The rule's conditions, such as which repository or branch may ask, decide who can obtain the role. ## An example, sourced GitHub's documentation says OpenID Connect allows workflows to exchange short-lived tokens directly from the cloud provider (read 16 Nov 2026). ## Why it matters here On Cloud Security Breaches the term appears in the rubric and on the platform pages; the example above was read on 16 Nov 2026 from the page under Sources. ## Related pages - [Session token: meaning and an example](/glossary/session-token/) - [OAuth token: meaning and an example](/glossary/oauth-token/) - [Infostealer: meaning and an example](/glossary/infostealer/) - [All of them](/glossary/) ## Sources - https://docs.github.com/en/actions/security-for-github-actions/security-hardening-your-deployments/about-security-hardening-with-openid-connect (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/glossary/oidc-trust/ --- # Session token Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. A session token is the proof of a completed sign-in that a browser or application presents instead of a password. A copy stolen from a device can be replayed to enter an account without a new sign-in, sometimes including a second factor already satisfied. ## An example, sourced Google Cloud's M-Trends 2026 post says threat actors are bypassing standard defenses by harvesting long-lived OAuth tokens and session cookies (read 16 Nov 2026). ## Why it matters here On Cloud Security Breaches the term appears in the rubric and on the platform pages; the example above was read on 16 Nov 2026 from the page under Sources. ## Related pages - [OAuth token: meaning and an example](/glossary/oauth-token/) - [Infostealer: meaning and an example](/glossary/infostealer/) - [Multi-factor authentication: meaning and an example](/glossary/multi-factor-authentication/) - [All of them](/glossary/) ## Sources - https://cloud.google.com/blog/topics/threat-intelligence/m-trends-2026 (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/glossary/session-token/ --- # OAuth token Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. An OAuth token is a credential that lets an application act for a user or a company in another service without holding a password. Tokens held by an integration can be reused from anywhere until they expire or are revoked. ## An example, sourced The M-Trends 2026 post from Google Cloud says attackers steal hard-coded keys and personal access tokens by compromising third-party SaaS vendors (read 16 Nov 2026). ## Why it matters here On Cloud Security Breaches the term appears in the rubric and on the platform pages; the example above was read on 16 Nov 2026 from the page under Sources. ## Related pages - [Infostealer: meaning and an example](/glossary/infostealer/) - [Multi-factor authentication: meaning and an example](/glossary/multi-factor-authentication/) - [Least privilege: meaning and an example](/glossary/least-privilege/) - [All of them](/glossary/) ## Sources - https://cloud.google.com/blog/topics/threat-intelligence/m-trends-2026 (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/glossary/oauth-token/ --- # Infostealer Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. An infostealer is malware that collects saved passwords, browser cookies and session tokens from an infected device and sends them to its operator. The stolen items are then used to sign in to cloud and software-as-a-service accounts as the device's owner. ## An example, sourced Google Cloud's Cloud Threat Horizons report for the first half of 2026 is one provider publication that tracks credential theft against cloud accounts (read 16 Nov 2026). ## Why it matters here On Cloud Security Breaches the term appears in the rubric and on the platform pages; the example above was read on 16 Nov 2026 from the page under Sources. ## Related pages - [Multi-factor authentication: meaning and an example](/glossary/multi-factor-authentication/) - [Least privilege: meaning and an example](/glossary/least-privilege/) - [Permissions boundary: meaning and an example](/glossary/permissions-boundary/) - [All of them](/glossary/) ## Sources - https://cloud.google.com/blog/topics/threat-intelligence/cloud-threat-horizons-report-h1-2026 (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/glossary/infostealer/ --- # Multi-factor authentication Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Multi-factor authentication asks for a second proof of identity, such as a code or a hardware key, in addition to a password. A stolen password alone then fails, although a stolen session token or a tricked user can still get through some forms of it. ## An example, sourced Microsoft's Entra documentation explains how multi-factor authentication works and which verification methods it offers (read 16 Nov 2026). ## Why it matters here On Cloud Security Breaches the term appears in the rubric and on the platform pages; the example above was read on 16 Nov 2026 from the page under Sources. ## Related pages - [Least privilege: meaning and an example](/glossary/least-privilege/) - [Permissions boundary: meaning and an example](/glossary/permissions-boundary/) - [Unused credential: meaning and an example](/glossary/unused-credential/) - [All of them](/glossary/) ## Sources - https://learn.microsoft.com/en-us/entra/identity/authentication/concept-mfa-howitworks (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/glossary/multi-factor-authentication/ --- # Least privilege Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Least privilege is the practice of giving each identity only the actions on the resources that its task needs. In an incident it limits what a stolen credential can reach, because the credential carries no more access than its owner had. ## An example, sourced AWS's IAM best practices page lists applying least-privilege permissions as one of its recommendations (read 16 Nov 2026). ## Why it matters here On Cloud Security Breaches the term appears in the rubric and on the platform pages; the example above was read on 16 Nov 2026 from the page under Sources. ## Related pages - [Permissions boundary: meaning and an example](/glossary/permissions-boundary/) - [Unused credential: meaning and an example](/glossary/unused-credential/) - [Offboarding: meaning and an example](/glossary/offboarding/) - [All of them](/glossary/) ## Sources - https://docs.aws.amazon.com/IAM/latest/UserGuide/best-practices.html (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/glossary/least-privilege/ --- # Permissions boundary Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. A permissions boundary is a policy attached to an identity that sets the most that its other policies can ever grant. Even if someone adds a broad permission to the identity later, the boundary caps what the identity can do. ## An example, sourced AWS's documentation says a permissions boundary uses a managed policy to set the maximum permissions that an identity-based policy can grant to an IAM entity (read 16 Nov 2026). ## Why it matters here On Cloud Security Breaches the term appears in the rubric and on the platform pages; the example above was read on 16 Nov 2026 from the page under Sources. ## Related pages - [Unused credential: meaning and an example](/glossary/unused-credential/) - [Offboarding: meaning and an example](/glossary/offboarding/) - [Instance metadata service: meaning and an example](/glossary/instance-metadata-service/) - [All of them](/glossary/) ## Sources - https://docs.aws.amazon.com/IAM/latest/UserGuide/access_policies_boundaries.html (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/glossary/permissions-boundary/ --- # Unused credential Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. An unused credential is a password, key or token that has not been used for a long period but still works. It stays valid for anyone who finds it, and it gives an intruder access that no current owner watches. ## An example, sourced AWS publishes a page on finding unused AWS credentials, which describes reports and last-used data for passwords and keys (read 16 Nov 2026). ## Why it matters here On Cloud Security Breaches the term appears in the rubric and on the platform pages; the example above was read on 16 Nov 2026 from the page under Sources. ## Related pages - [Offboarding: meaning and an example](/glossary/offboarding/) - [Instance metadata service: meaning and an example](/glossary/instance-metadata-service/) - [Server-side request forgery: meaning and an example](/glossary/server-side-request-forgery/) - [All of them](/glossary/) ## Sources - https://docs.aws.amazon.com/IAM/latest/UserGuide/id_credentials_finding-unused.html (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/glossary/unused-credential/ --- # Offboarding Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Offboarding is the removal of a leaving person's or contractor's access to every account, key and connected application they held. Retained-access incidents begin where this removal is incomplete, for example a key that was never rotated. The ledger records such cases under the retained-access path. ## An example, sourced The ledger's retained-access path covers incidents where a former employee or contractor kept access; AWS's access key documentation describes the rotation and deletion that removal needs (read 16 Nov 2026). ## Why it matters here On Cloud Security Breaches the term appears in the rubric and on the platform pages; the example above was read on 16 Nov 2026 from the page under Sources. ## Related pages - [Instance metadata service: meaning and an example](/glossary/instance-metadata-service/) - [Server-side request forgery: meaning and an example](/glossary/server-side-request-forgery/) - [Security group: meaning and an example](/glossary/security-group/) - [All of them](/glossary/) ## Sources - https://docs.aws.amazon.com/IAM/latest/UserGuide/id_credentials_access-keys.html (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/glossary/offboarding/ --- # Instance metadata service Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. The instance metadata service is a local address that a virtual machine can query for facts about itself, including temporary credentials for the role attached to it. Any software on the machine, or any request forged to reach that address, can read what it returns. ## An example, sourced AWS's documentation describes the Instance Metadata Service, with a session-oriented version (IMDSv2) and an earlier version (IMDSv1) (read 16 Nov 2026). ## Why it matters here On Cloud Security Breaches the term appears in the rubric and on the platform pages; the example above was read on 16 Nov 2026 from the page under Sources. ## Related pages - [Server-side request forgery: meaning and an example](/glossary/server-side-request-forgery/) - [Security group: meaning and an example](/glossary/security-group/) - [Network security group: meaning and an example](/glossary/network-security-group/) - [All of them](/glossary/) ## Sources - https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/configuring-instance-metadata-service.html (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/glossary/instance-metadata-service/ --- # Server-side request forgery Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Server-side request forgery, or SSRF, is a flaw that makes an application send a request to an address the attacker picks. Inside a cloud, that address can be the metadata service, which returns credentials belonging to the machine's role. Providers have added session-based metadata requests to blunt it. ## An example, sourced AWS's security blog describes defense in depth against open firewalls, reverse proxies and SSRF vulnerabilities through enhancements to the EC2 Instance Metadata Service (read 16 Nov 2026). ## Why it matters here On Cloud Security Breaches the term appears in the rubric and on the platform pages; the example above was read on 16 Nov 2026 from the page under Sources. ## Related pages - [Security group: meaning and an example](/glossary/security-group/) - [Network security group: meaning and an example](/glossary/network-security-group/) - [Block public access: meaning and an example](/glossary/block-public-access/) - [All of them](/glossary/) ## Sources - https://aws.amazon.com/blogs/security/defense-in-depth-open-firewalls-reverse-proxies-ssrf-vulnerabilities-ec2-instance-metadata-service/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/glossary/server-side-request-forgery/ --- # Security group Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. A security group is a set of allow rules that controls which network traffic can reach a cloud resource and from where. A rule that admits any address on a management or database port makes the service reachable from the whole internet. ## An example, sourced AWS publishes a user guide for VPC security groups, which sets out how inbound and outbound rules apply to resources (read 16 Nov 2026). ## Why it matters here On Cloud Security Breaches the term appears in the rubric and on the platform pages; the example above was read on 16 Nov 2026 from the page under Sources. ## Related pages - [Network security group: meaning and an example](/glossary/network-security-group/) - [Block public access: meaning and an example](/glossary/block-public-access/) - [Public access prevention: meaning and an example](/glossary/public-access-prevention/) - [All of them](/glossary/) ## Sources - https://docs.aws.amazon.com/vpc/latest/userguide/security-groups.html (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/glossary/security-group/ --- # Network security group Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. A network security group is Azure's rule set for allowing or denying inbound and outbound traffic to resources in a virtual network. Its rules, like those of any firewall, decide whether a service is reachable from outside. A rule that admits all addresses to a management port exposes that service to the internet. ## An example, sourced Microsoft's documentation says a network security group contains security rules that allow or deny inbound or outbound network traffic to several types of Azure resources (read 16 Nov 2026). ## Why it matters here On Cloud Security Breaches the term appears in the rubric and on the platform pages; the example above was read on 16 Nov 2026 from the page under Sources. ## Related pages - [Block public access: meaning and an example](/glossary/block-public-access/) - [Public access prevention: meaning and an example](/glossary/public-access-prevention/) - [Anonymous read access: meaning and an example](/glossary/anonymous-read-access/) - [All of them](/glossary/) ## Sources - https://learn.microsoft.com/en-us/azure/virtual-network/network-security-groups-overview (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/glossary/network-security-group/ --- # Block public access Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Block public access is an account or bucket setting in Amazon S3 that overrides policies and permissions which would otherwise allow anonymous access to stored objects. It turns a later mistaken grant into a request that is refused. It works at the account level, so one switch covers every bucket in that account. ## An example, sourced AWS's documentation says new buckets, access points and objects do not allow public access by default, but that users can modify policies or permissions to allow it (read 16 Nov 2026). ## Why it matters here On Cloud Security Breaches the term appears in the rubric and on the platform pages; the example above was read on 16 Nov 2026 from the page under Sources. ## Related pages - [Public access prevention: meaning and an example](/glossary/public-access-prevention/) - [Anonymous read access: meaning and an example](/glossary/anonymous-read-access/) - [Shared access signature: meaning and an example](/glossary/shared-access-signature/) - [All of them](/glossary/) ## Sources - https://docs.aws.amazon.com/AmazonS3/latest/userguide/access-control-block-public-access.html (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/glossary/block-public-access/ --- # Public access prevention Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Public access prevention is a Google Cloud Storage setting that stops a bucket or a whole organization from granting access to anonymous users. It acts as a backstop when an individual permission is set too widely. It can be enforced across an organization, so a single project owner cannot undo it alone. ## An example, sourced Google Cloud publishes a Cloud Storage documentation page titled Public access prevention (read 16 Nov 2026). ## Why it matters here On Cloud Security Breaches the term appears in the rubric and on the platform pages; the example above was read on 16 Nov 2026 from the page under Sources. ## Related pages - [Anonymous read access: meaning and an example](/glossary/anonymous-read-access/) - [Shared access signature: meaning and an example](/glossary/shared-access-signature/) - [Tenant: meaning and an example](/glossary/tenant/) - [All of them](/glossary/) ## Sources - https://cloud.google.com/storage/docs/public-access-prevention (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/glossary/public-access-prevention/ --- # Anonymous read access Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Anonymous read access lets anyone with a container's address download its contents without signing in. It is the setting behind many exposed-storage incidents, and cloud providers offer an account-level control to block it. Without it, a container set to public stays readable by anyone who learns its web address. ## An example, sourced Microsoft's documentation describes how to prevent anonymous read access to containers and blobs in Azure Storage (read 16 Nov 2026). ## Why it matters here On Cloud Security Breaches the term appears in the rubric and on the platform pages; the example above was read on 16 Nov 2026 from the page under Sources. ## Related pages - [Shared access signature: meaning and an example](/glossary/shared-access-signature/) - [Tenant: meaning and an example](/glossary/tenant/) - [Multitenant application: meaning and an example](/glossary/multitenant-application/) - [All of them](/glossary/) ## Sources - https://learn.microsoft.com/en-us/azure/storage/blobs/anonymous-read-access-prevent (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/glossary/anonymous-read-access/ --- # Shared access signature Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. A shared access signature is a signed link that gives a client delegated access to specific Azure Storage resources, with set permissions and a set expiry. A link created with a distant expiry and broad permissions works for anyone who holds it. ## An example, sourced Microsoft's documentation says a shared access signature provides secure delegated access, with control over which resources, which permissions and how long the signature is valid (read 16 Nov 2026). ## Why it matters here On Cloud Security Breaches the term appears in the rubric and on the platform pages; the example above was read on 16 Nov 2026 from the page under Sources. ## Related pages - [Tenant: meaning and an example](/glossary/tenant/) - [Multitenant application: meaning and an example](/glossary/multitenant-application/) - [Confused deputy: meaning and an example](/glossary/confused-deputy/) - [All of them](/glossary/) ## Sources - https://learn.microsoft.com/en-us/azure/storage/common/storage-sas-overview (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/glossary/shared-access-signature/ --- # Tenant Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. A tenant is the boundary a provider draws around one customer's identities, applications and data inside a shared service. Cross-tenant incidents are those where something crossed that boundary, through a provider flaw or a customer setting. Each tenant holds its own users, policies and registered applications, apart from every other customer. ## An example, sourced Microsoft's documentation says Entra ID organizes objects like users and apps into groups called tenants, which let an administrator set policies for the organization (read 16 Nov 2026). ## Why it matters here On Cloud Security Breaches the term appears in the rubric and on the platform pages; the example above was read on 16 Nov 2026 from the page under Sources. ## Related pages - [Multitenant application: meaning and an example](/glossary/multitenant-application/) - [Confused deputy: meaning and an example](/glossary/confused-deputy/) - [External ID: meaning and an example](/glossary/external-id/) - [All of them](/glossary/) ## Sources - https://learn.microsoft.com/en-us/entra/identity-platform/single-and-multi-tenant-apps (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/glossary/tenant/ --- # Multitenant application Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. A multitenant application is software registered to accept sign-ins from users in many organizations' tenants. Its configuration decides which tenants may sign in and what each can reach, so a loose setting can open it to accounts from any tenant. ## An example, sourced Microsoft publishes a page on single and multitenant apps in Entra ID, covering how an app can be limited to one tenant or open to many (read 16 Nov 2026). ## Why it matters here On Cloud Security Breaches the term appears in the rubric and on the platform pages; the example above was read on 16 Nov 2026 from the page under Sources. ## Related pages - [Confused deputy: meaning and an example](/glossary/confused-deputy/) - [External ID: meaning and an example](/glossary/external-id/) - [Third-party integration: meaning and an example](/glossary/third-party-integration/) - [All of them](/glossary/) ## Sources - https://learn.microsoft.com/en-us/entra/identity-platform/single-and-multi-tenant-apps (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/glossary/multitenant-application/ --- # Confused deputy Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. The confused deputy problem is a flaw where a party without permission to act tricks a more privileged party into acting for it. In cloud it arises when a service or vendor role is invoked on behalf of a customer who is not the intended one. ## An example, sourced AWS's documentation says the confused deputy problem is a security issue where an entity that lacks permission to perform an action can coerce a more-privileged entity to perform it (read 16 Nov 2026). ## Why it matters here On Cloud Security Breaches the term appears in the rubric and on the platform pages; the example above was read on 16 Nov 2026 from the page under Sources. ## Related pages - [External ID: meaning and an example](/glossary/external-id/) - [Third-party integration: meaning and an example](/glossary/third-party-integration/) - [Form 8-K Item 1.05: meaning and an example](/glossary/form-8-k-item-1-05/) - [All of them](/glossary/) ## Sources - https://docs.aws.amazon.com/IAM/latest/UserGuide/confused-deputy.html (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/glossary/confused-deputy/ --- # External ID Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. An external ID is an identifier placed in a role's trust policy so that a vendor assuming the role must present the value agreed for one customer. It prevents the vendor's access from being used on behalf of a different customer. ## An example, sourced AWS's documentation on third-party access describes using an external ID when you allow a third party to assume a role in your account (read 16 Nov 2026). ## Why it matters here On Cloud Security Breaches the term appears in the rubric and on the platform pages; the example above was read on 16 Nov 2026 from the page under Sources. ## Related pages - [Third-party integration: meaning and an example](/glossary/third-party-integration/) - [Form 8-K Item 1.05: meaning and an example](/glossary/form-8-k-item-1-05/) - [Consent order: meaning and an example](/glossary/consent-order/) - [All of them](/glossary/) ## Sources - https://docs.aws.amazon.com/IAM/latest/UserGuide/id_roles_common-scenarios_third-party.html (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/glossary/external-id/ --- # Third-party integration Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. A third-party integration is an outside application connected to a company's cloud or software-as-a-service account by a token, key or role. Its access lasts as long as the grant does, whatever happens in the vendor's own systems. The ledger records incidents that began in an integration as third-party access, not as the customer's own compromise. ## An example, sourced Microsoft's Entra documentation covers managing application permissions, including the consent granted to applications in a tenant (read 16 Nov 2026). ## Why it matters here On Cloud Security Breaches the term appears in the rubric and on the platform pages; the example above was read on 16 Nov 2026 from the page under Sources. ## Related pages - [Form 8-K Item 1.05: meaning and an example](/glossary/form-8-k-item-1-05/) - [Consent order: meaning and an example](/glossary/consent-order/) - [Provider post-mortem: meaning and an example](/glossary/provider-post-mortem/) - [All of them](/glossary/) ## Sources - https://learn.microsoft.com/en-us/entra/identity/enterprise-apps/manage-application-permissions (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/glossary/third-party-integration/ --- # Form 8-K Item 1.05 Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Form 8-K Item 1.05 is the section of the U.S. current report in which a listed company describes a material cybersecurity incident. The ledger treats such a filing as a company statement and quotes only what the filing says. A filing may name no provider or access path, in which case the ledger records both as not public. ## An example, sourced A Nutex Health Form 8-K dated 31 August 2026 carries the heading Item 1.05 Material Cybersecurity Incidents (read 16 Nov 2026). ## Why it matters here On Cloud Security Breaches the term appears in the rubric and on the platform pages; the example above was read on 16 Nov 2026 from the page under Sources. ## Related pages - [Consent order: meaning and an example](/glossary/consent-order/) - [Provider post-mortem: meaning and an example](/glossary/provider-post-mortem/) - [Access path: meaning and an example](/glossary/access-path/) - [All of them](/glossary/) ## Sources - https://www.sec.gov/Archives/edgar/data/1479681/000162828026059602/nutx-20260831.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/glossary/form-8-k-item-1-05/ --- # Consent order Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. A consent order is a regulator's order that a company agrees to without admitting or denying the findings. It sets out required steps and sometimes a penalty, and it counts in the ledger as a regulator filing. It also fixes the date and the findings that the ledger can quote verbatim from the order itself. ## An example, sourced The U.S. Office of the Comptroller of the Currency published a consent order for the assessment of a civil money penalty, number ea2020-036 (read 16 Nov 2026). ## Why it matters here On Cloud Security Breaches the term appears in the rubric and on the platform pages; the example above was read on 16 Nov 2026 from the page under Sources. ## Related pages - [Provider post-mortem: meaning and an example](/glossary/provider-post-mortem/) - [Access path: meaning and an example](/glossary/access-path/) - [Access key: meaning and an example](/glossary/access-key/) - [All of them](/glossary/) ## Sources - https://www.occ.gov/static/enforcement-actions/ea2020-036.pdf (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/glossary/consent-order/ --- # Provider post-mortem Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. A provider post-mortem is a cloud or software-as-a-service provider's own written account of an incident on its platform, covering the timeline, cause and fix. The ledger accepts these as primary sources and quotes them with the provider's wording. A post-mortem that names no cause leaves that field as not public in the ledger. ## An example, sourced AWS publishes summaries of service events on its message page, and Microsoft publishes Azure post-incident reviews on its status history page (read 16 Nov 2026). ## Why it matters here On Cloud Security Breaches the term appears in the rubric and on the platform pages; the example above was read on 16 Nov 2026 from the page under Sources. ## Related pages - [Access path: meaning and an example](/glossary/access-path/) - [Access key: meaning and an example](/glossary/access-key/) - [Temporary credentials: meaning and an example](/glossary/temporary-credentials/) - [All of them](/glossary/) ## Sources - https://aws.amazon.com/message/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/glossary/provider-post-mortem/ --- # Access path Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. An access path is the route an intruder used to reach cloud data, as the primary sources state it. The ledger sorts every entry into one of eight paths, such as leaked keys or third-party access, and shows a second path only when the sources do. ## An example, sourced The ledger's eight paths are described, with provider documentation, on the access paths page; this entry uses the AWS IAM best practices page for the controls that match them (read 16 Nov 2026). ## Why it matters here On Cloud Security Breaches the term appears in the rubric and on the platform pages; the example above was read on 16 Nov 2026 from the page under Sources. ## Related pages - [Access key: meaning and an example](/glossary/access-key/) - [Temporary credentials: meaning and an example](/glossary/temporary-credentials/) - [Service account key: meaning and an example](/glossary/service-account-key/) - [All of them](/glossary/) ## Sources - https://docs.aws.amazon.com/IAM/latest/UserGuide/best-practices.html (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/glossary/access-path/ --- # What changed in the last 90 days Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Dated items from primary sources: incident notices from company filings, regulator actions and provider post-mortems, from 18 August 2026 to 16 November 2026. Each item names the company or provider that published it, quotes the source where it is brief and links to the original document read on 16 November 2026. - [Cloudflare reports cross-request data exposure in a Workers AI model](/news/cloudflare-workers-ai-kv-cache-exposure/): 26 Oct 2026 - [Cloudflare reports eight compromised intermediate certificate keys](/news/cloudflare-emudhra-certificates/): 20 Oct 2026 - [Park Dental Partners confirms data acquisition and four lawsuits](/news/park-dental-partners-lawsuits/): 13 Oct 2026 - [Microsoft publishes review of Azure gateway control plane failure](/news/azure-gateway-control-plane-pir/): 30 Sep 2026 - [Microsoft publishes review of Azure OpenAI Sweden Central outage](/news/azure-openai-sweden-central-pir/): 29 Sep 2026 - [Astrana Health files Item 1.05 after social engineering attempts](/news/astrana-health-social-engineering/): 23 Sep 2026 - [Nutex Health says the third party published the data](/news/nutex-health-data-published/): 11 Sep 2026 - [Boston Scientific says its August incident is likely material](/news/boston-scientific-material-impact/): 8 Sep 2026 - [Veradigm files Item 1.05 on credentials taken from a vendor's environment](/news/veradigm-vendor-api-2026/): 8 Sep 2026 - [Park Dental Partners reports unauthorized access to its network](/news/park-dental-partners-unauthorized-access/): 1 Sep 2026 - [Nutex Health files Item 1.05 and discloses a class action](/news/nutex-health-material-incident/): 31 Aug 2026 - [Nutex Health reports unauthorized activity on its network](/news/nutex-health-unauthorized-activity/): 24 Aug 2026 ## Related pages - [The eight access paths in cloud breaches](/access-paths/) - [Cloud breaches by cloud and provider](/clouds/) - [Cloud breaches by year, 2019 to 2026](/years/) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/news/ --- # Cloudflare reports cross-request data exposure in a Workers AI model Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Cloudflare's 26 October 2026 post says a configuration mismatch in one hosted model exposed data across customers' requests between 27 August and 9 September 2026. The company states 155 responses were affected and that it found no evidence of deliberate exploitation. Cloudflare states that a small number of customers using the GLM 5.3 model on Workers AI received responses containing data from other customers' requests made shortly before theirs, all for prompts longer than 681,280 tokens. The post says the issue was reported on 8 September through its bug bounty program, reproduced the next morning, and the model was taken offline within 40 minutes of declaring an incident and restored on 9 September. It says no other Workers AI model was affected. Cloudflare published the report alongside a post from the maintainers of the open source inference engine it uses. The ledger would class this as a cross-tenant exposure at the provider. ## Related pages - [Cloudflare reports eight compromised intermediate certificate keys](/news/cloudflare-emudhra-certificates/) - [Park Dental Partners confirms data acquisition and four lawsuits](/news/park-dental-partners-lawsuits/) - [Microsoft publishes review of Azure gateway control plane failure](/news/azure-gateway-control-plane-pir/) - [All of them](/news/) ## Sources - https://blog.cloudflare.com/kv-cache-exposure-remediation/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/news/cloudflare-workers-ai-kv-cache-exposure/ --- # Cloudflare reports eight compromised intermediate certificate keys Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Cloudflare's blog post of 20 October 2026 says it learned on 19 October 2026 that private keys for eight eMudhra intermediate certificates were compromised. Cloudflare states its security review found no active exploitation on its services and lists client updates and mitigations across its products. The post says the certificates were widely trusted by browsers, operating systems and within Cloudflare, and that because the keys are weak an attacker can forge a certificate for any domain. Cloudflare states it released updates for its device client and tunnel software and deployed mitigations across its CDN, Developer and Zero Trust products, and that some customers must act on trust configurations they operate, especially on origin servers. The post describes a certificate trust problem at a certificate authority, not a customer cloud account, and is logged here as a provider post. ## Related pages - [Park Dental Partners confirms data acquisition and four lawsuits](/news/park-dental-partners-lawsuits/) - [Microsoft publishes review of Azure gateway control plane failure](/news/azure-gateway-control-plane-pir/) - [Microsoft publishes review of Azure OpenAI Sweden Central outage](/news/azure-openai-sweden-central-pir/) - [All of them](/news/) ## Sources - https://blog.cloudflare.com/emudhra-certificate-compromise/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/news/cloudflare-emudhra-certificates/ --- # Park Dental Partners confirms data acquisition and four lawsuits Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Park Dental Partners, Inc. stated in an 8-K dated 13 October 2026 that its investigation determined data was subject to unauthorized acquisition. The company reported four putative class actions in Ramsey County, Minnesota, the first filed on 3 September 2026, and a consolidation motion filed on 7 October 2026. The filing says the company identified impacted patients whose protected health information was involved and is completing notifications under state and federal law. It states that the lawsuits allege failure to safeguard protected health information or to implement reasonable security measures, and failure to notify affected individuals in time. The company stated it intends to defend itself and anticipates regulatory responses. The company said it was not aware of any unauthorized use of protected health information. The filing does not describe the access path or name a cloud provider. ## Related pages - [Microsoft publishes review of Azure gateway control plane failure](/news/azure-gateway-control-plane-pir/) - [Microsoft publishes review of Azure OpenAI Sweden Central outage](/news/azure-openai-sweden-central-pir/) - [Astrana Health files Item 1.05 after social engineering attempts](/news/astrana-health-social-engineering/) - [All of them](/news/) ## Sources - https://www.sec.gov/Archives/edgar/data/2069604/000110465926115541/park-20261007x8k.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/news/park-dental-partners-lawsuits/ --- # Microsoft publishes review of Azure gateway control plane failure Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Microsoft's post incident review for 30 September 2026 says gateway updates resumed across many regions at once and overwhelmed a regional networking control plane. Management operations failed intermittently from 20:30 UTC on 30 September to 02:15 UTC on 1 October 2026 across more than thirty regions. The review names Azure ExpressRoute Gateway, Application Gateway, Firewall, Bastion, Virtual WAN and VPN Gateway among the affected services, and says some upgraded ExpressRoute gateways temporarily lost connectivity. Microsoft states that a temporary deployment restriction expired, letting a large number of gateway updates resume within a short period, while an earlier change had lengthened configuration processing time on a shared control plane service. The recovery step was a rollback of that earlier change, started at 01:15 UTC on 1 October 2026. Microsoft lists work to make paused updates resume in staged rollouts. The review describes an availability incident. ## Related pages - [Microsoft publishes review of Azure OpenAI Sweden Central outage](/news/azure-openai-sweden-central-pir/) - [Astrana Health files Item 1.05 after social engineering attempts](/news/astrana-health-social-engineering/) - [Nutex Health says the third party published the data](/news/nutex-health-data-published/) - [All of them](/news/) ## Sources - https://azure.status.microsoft/en-us/status/history/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/news/azure-gateway-control-plane-pir/ --- # Microsoft publishes review of Azure OpenAI Sweden Central outage Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Microsoft's Azure post incident review for 29 September 2026 says a planned deployment cleared an internal service's caches and sent backend traffic to about 20 times normal. Azure OpenAI Service and related services in Sweden Central had intermittent failures from 10:03 to 15:58 UTC. The review says two other defects in how the internal service read from its caches caused most lookups to bypass the cache and call the backend service directly. Connection capacity on infrastructure nodes was then exhausted. Microsoft states that the extended duration was primarily caused by health checks that depended on connecting to the database layer, which put instances into a restart loop. Service availability began to recover at 15:40 UTC after the health check was reconfigured. The review lists repairs including throttling and rate limiting for the backend service, more caching capacity and better diagnostic metrics. It describes an availability incident. ## Related pages - [Astrana Health files Item 1.05 after social engineering attempts](/news/astrana-health-social-engineering/) - [Nutex Health says the third party published the data](/news/nutex-health-data-published/) - [Boston Scientific says its August incident is likely material](/news/boston-scientific-material-impact/) - [All of them](/news/) ## Sources - https://azure.status.microsoft/en-us/status/history/ (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/news/azure-openai-sweden-central-pir/ --- # Astrana Health files Item 1.05 after social engineering attempts Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Astrana Health, Inc. disclosed on 23 September 2026 that threat actors impersonating company personnel and spoofing its main corporate telephone number contacted employees to obtain unauthorized access. The company determined the incident material as of 22 September 2026 and listed credential resets and remote access restrictions among its remedial steps. The Form 8-K says the company's cybersecurity team detected and responded to the activity, engaged a third-party forensic firm, notified law enforcement and is notifying state and federal regulators and payer partners. Remedial measures listed in the filing include resetting affected credentials, restricting remote access tools, restoring certain systems from clean backups and enhancing monitoring, logging and detection. The company stated that it believed certain private or confidential information on its servers had been accessed or acquired without authorization. The filing does not name a cloud provider. ## Related pages - [Nutex Health says the third party published the data](/news/nutex-health-data-published/) - [Boston Scientific says its August incident is likely material](/news/boston-scientific-material-impact/) - [Veradigm files Item 1.05 on credentials taken from a vendor's environment](/news/veradigm-vendor-api-2026/) - [All of them](/news/) ## Sources - https://www.sec.gov/Archives/edgar/data/1083446/000110465926109813/asth-20260922x8k.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/news/astrana-health-social-engineering/ --- # Nutex Health says the third party published the data Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Nutex Health Inc. stated in an 8-K dated 11 September 2026 that it had learned the unauthorized third party published on its website the data allegedly obtained. The company said the review of the volume of data would take several weeks and that its materiality assessment had not changed. The filing says the company, with its cybersecurity advisors, was downloading, processing and analyzing the published data to determine its contents, scope and authenticity. It states that several purported class action complaints had been filed in the Southern District of Texas, and that the company was unable to predict the outcome or estimate the impact on its business. The company said it continued to evaluate notification requirements and intended to notify impacted patients and employees. The filing gives no access path. ## Related pages - [Boston Scientific says its August incident is likely material](/news/boston-scientific-material-impact/) - [Veradigm files Item 1.05 on credentials taken from a vendor's environment](/news/veradigm-vendor-api-2026/) - [Park Dental Partners reports unauthorized access to its network](/news/park-dental-partners-unauthorized-access/) - [All of them](/news/) ## Sources - https://www.sec.gov/Archives/edgar/data/1479681/000162828026061432/materiality_assessmentxupd.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/news/nutex-health-data-published/ --- # Boston Scientific says its August incident is likely material Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Boston Scientific Corporation filed an Item 1.05 Form 8-K on 8 September 2026 saying the incident it identified on 25 August 2026 is likely to have a material impact on third quarter and full year 2026 results. The company said it had not identified evidence of ongoing unauthorized access. The filing says the incident involved unauthorized activity on certain systems and caused a network outage that affected the company's ability to manufacture, process and ship customer orders. The company stated that distribution centers were processing and shipping orders at or above normal operating levels and that manufacturing had resumed across most facilities globally, while the timeline for full recovery was uncertain. The filing says the company believes it is unlikely to meet its third quarter and full year 2026 net sales growth and adjusted EPS guidance ranges. It does not state how access was gained or name a cloud provider. ## Related pages - [Veradigm files Item 1.05 on credentials taken from a vendor's environment](/news/veradigm-vendor-api-2026/) - [Park Dental Partners reports unauthorized access to its network](/news/park-dental-partners-unauthorized-access/) - [Nutex Health files Item 1.05 and discloses a class action](/news/nutex-health-material-incident/) - [All of them](/news/) ## Sources - https://www.sec.gov/Archives/edgar/data/885725/000088572526000059/bsx-20260907.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/news/boston-scientific-material-impact/ --- # Veradigm files Item 1.05 on credentials taken from a vendor's environment Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Veradigm Inc. filed a Form 8-K on 8 September 2026 saying an unauthorized party took credentials to a company application programming interface from a third-party vendor's environment and used them to download patient personal data, including some Social Security numbers. The filing says no clinical data was involved; the ledger records it under third-party access. The filing says a third-party vendor had an incident, and that credentials obtained from the vendor's environment gave access to a company application programming interface, through which copies of certain patient personal data were downloaded. The filing says a small number of customers were affected and that no clinical or medical data was involved. The cloud or hosting provider is not named, so that field reads not public in the [ledger entry](/incidents/veradigm-vendor-api-2026/). ## Related pages - [Park Dental Partners reports unauthorized access to its network](/news/park-dental-partners-unauthorized-access/) - [Nutex Health files Item 1.05 and discloses a class action](/news/nutex-health-material-incident/) - [Nutex Health reports unauthorized activity on its network](/news/nutex-health-unauthorized-activity/) - [All of them](/news/) ## Sources - https://www.sec.gov/Archives/edgar/data/1124804/000119312526385249/mdrx-20260908.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/news/veradigm-vendor-api-2026/ --- # Park Dental Partners reports unauthorized access to its network Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Park Dental Partners, Inc. filed an Item 1.05 Form 8-K dated 1 September 2026 stating that it identified unauthorized access to its computer network on 28 August 2026. The company said it was treating the event as reportable because patient data may have been accessed. The filing names no cloud provider. The Form 8-K says the company promptly started its incident response protocols and engaged external cybersecurity and forensic specialists. It states that the incident had not materially disrupted operations. The company said it could not yet estimate the financial, legal, operational and reputational impact and that it had not identified any material impact on its financial condition, results of operations or business operations. The filing does not describe how access was gained. ## Related pages - [Nutex Health files Item 1.05 and discloses a class action](/news/nutex-health-material-incident/) - [Nutex Health reports unauthorized activity on its network](/news/nutex-health-unauthorized-activity/) - [Ledger additions, edition by edition](/editions/) - [All of them](/news/) ## Sources - https://www.sec.gov/Archives/edgar/data/2069604/000110465926104300/park-20260828x8k.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/news/park-dental-partners-unauthorized-access/ --- # Nutex Health files Item 1.05 and discloses a class action Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Nutex Health Inc. filed an Item 1.05 Form 8-K on 31 August 2026 saying patient, employee, credentialed provider and business information was accessed and exfiltrated, and that the third party threatened to post it. A purported class action was filed in the Southern District of Texas on 27 August 2026. The filing updates the company's 24 August 2026 Item 8.01 report. It says the third party has threatened to post the information externally, and that the company had not identified any material impact on its business operations or financial reporting systems. It also describes a purported class action, Case No. 4:26-cv-07197 against Nutex Health, Inc., filed on 27 August 2026 in the United States District Court for the Southern District of Texas, Houston Division, asserting negligence, negligence per se, breach of third-party beneficiary contract and unjust enrichment. The filing gives no access path and names no cloud provider. ## Related pages - [Nutex Health reports unauthorized activity on its network](/news/nutex-health-unauthorized-activity/) - [Ledger additions, edition by edition](/editions/) - [Ledger edition 1: 111 incidents at launch](/editions/1/) - [All of them](/news/) ## Sources - https://www.sec.gov/Archives/edgar/data/1479681/000162828026059602/nutx-20260831.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/news/nutex-health-material-incident/ --- # Nutex Health reports unauthorized activity on its network Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Nutex Health Inc. told the SEC on 24 August 2026 that it learned of unauthorized activity involving data stored on its computer network. The company stated that it believes information on its servers was accessed and exfiltrated by an unauthorized third party. The filing does not name a cloud provider. Nutex Health's Form 8-K, filed under Item 8.01 on 24 August 2026, says the company engaged an independent third-party cybersecurity response team and forensic experts, activated a cybersecurity response plan, implemented containment measures and notified law enforcement. The filing says preliminary findings indicate that certain information maintained on the company's servers was accessed and exfiltrated by an unauthorized third party. The company stated that it had not identified any material impact on its business operations or financial reporting systems. The filing does not say which cloud or SaaS provider, if any, was involved, so the ledger records the cloud field as not public. ## Related pages - [Ledger additions, edition by edition](/editions/) - [Ledger edition 1: 111 incidents at launch](/editions/1/) - [Cloudflare reports cross-request data exposure in a Workers AI model](/news/cloudflare-workers-ai-kv-cache-exposure/) - [All of them](/news/) ## Sources - https://www.sec.gov/Archives/edgar/data/1479681/000162828026058606/nutx-20260824.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/news/nutex-health-unauthorized-activity/ --- # 30 questions, answered from the data Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Answers to common questions about the Cloud Breach Ledger: which sources it accepts, what not public means, how the eight access paths are assigned, how counts are computed and how to report an error. Each answer is short, names its source where one applies and gives the read date. **What is the Cloud Breach Ledger?** The Cloud Breach Ledger is a record of public cloud incidents from 2019 to 16 November 2026, with each entry read by the access path that was used. Every field comes from a company statement, regulator filing, court document or provider post-mortem. A field with no such source reads "not public". **Which sources does the ledger accept?** Company statements (press releases, blog posts, customer notices, SEC filings by the company such as an 8-K Item 1.05), regulator filings and decisions, court documents and provider post-mortems. News articles, vendor research blogs and Wikipedia are not accepted as sources. They may point to a primary source, nothing more. **What does "not public" mean in an entry?** It means no primary source states that fact as of the read date, 16 November 2026. It does not mean the fact is secret or unknown to the company. If a company later publishes it, the field changes and the correction is logged. **How many access paths does the ledger use?** Eight: leaked access keys, stolen user credentials, over-permissioned identities, public resources, cross-tenant flaws, third-party access, exposed network services and retained access. Each entry has one primary path and, only where the sources describe a second route, one secondary path. **How is an incident assigned to an access path?** The path follows how access was gained, in the words of the primary source. Where the source does not say how, the entry is assigned by what the source does say, and the access field reads "not public" for the rest. No path is inferred from rumor or from a news account. **Does the ledger rank companies or providers?** No. Entries are listed by date and filtered by path, provider or year. Counts by provider reflect how many incidents have a primary source that names that provider, so a higher count says nothing about which provider is safer. **Why does the ledger not name attackers or employees?** The ledger records organizations only. Where a company's own statement uses a group name, the ledger quotes it as the company's label. Otherwise it writes "an individual" or "a former employee", and it makes no claim about motive or blame. **Can an incident be in the ledger without a named cloud provider?** Yes, if the primary source describes cloud-hosted data or a cloud service but the provider field has no public statement behind it. The provider then reads "not public". The entry still qualifies when the source places the incident in a cloud or software-as-a-service tenant. **How are counts on the ledger computed?** Each count is the number of entries in the ledger that match a filter, for example one access path or one year of publication, on the edition date. They are counts of sourced incidents, not estimates of how many incidents happened. Entries added later change the next edition's counts. **What is the difference between a ledger count and a quoted statistic?** A ledger count is built from the entries on this site. A quoted statistic is a figure published by a third party, such as an annual breach report, and is shown with its publisher, year and link. The two are never added together or compared as if they measured the same thing. **How do I report an error in an entry?** Every entry has a corrections link to editor@cloudsecuritybreaches.com with the entry name in the subject. Send the primary source that shows the correct fact. A correction is accepted when a company statement, regulator filing, court document or provider post-mortem supports it. **How often is the ledger updated?** The ledger is published in dated editions, and this is the edition of 16 November 2026. The news page lists primary-source items dated 18 August 2026 onward. Each page shows the date its sources were read. **What is a leaked access key?** A leaked access key is a long-term key or token that ended up in a place others could read, such as a repository or a log, and was used. AWS's documentation says access keys are long-term credentials, and recommends [temporary credentials](https://docs.aws.amazon.com/IAM/latest/UserGuide/id_credentials_temp.html) in their place (read 16 Nov 2026). **What is the safer alternative to a long-term access key?** Temporary credentials that expire. AWS's [IAM best practices](https://docs.aws.amazon.com/IAM/latest/UserGuide/best-practices.html) page recommends that workloads use temporary credentials through IAM roles, and Google Cloud's documentation recommends choosing a more secure alternative to service account keys wherever possible (read 16 Nov 2026). **Why does a stolen password matter if the account has multi-factor authentication?** Because some second factors can be bypassed, and a stolen session token needs no second factor at all. Google Cloud's M-Trends 2026 post says threat actors harvest long-lived OAuth tokens and session cookies (read 16 Nov 2026). The control that applies is phishing-resistant sign-in and short session lifetimes. **What is an over-permissioned identity in an incident?** It is an identity whose access was wider than its job, so the extra access set how far the incident reached. The ledger uses this path when the primary source ties the scale of exposure to what the identity could reach, not only to how it was compromised. **How does a public resource differ from a leaked key?** A public resource needs no credential at all: anyone who finds the address can read it. A leaked key is a credential that someone used. Amazon S3's [Block Public Access](https://docs.aws.amazon.com/AmazonS3/latest/userguide/access-control-block-public-access.html) documentation says new buckets do not allow public access by default (read 16 Nov 2026). **What does cross-tenant mean in cloud incidents?** It means access crossed the boundary between two customers of the same provider. The cause can be a flaw in the provider's service or a setting in a customer's application. Microsoft's [tenant documentation](https://learn.microsoft.com/en-us/entra/identity-platform/single-and-multi-tenant-apps) explains how single and multitenant apps are configured (read 16 Nov 2026). **What counts as third-party access?** Access that arrived through a vendor, supplier, integration or contractor that the company had trusted. Examples in the ledger include incidents that began with tokens held by an integration. AWS's [confused deputy](https://docs.aws.amazon.com/IAM/latest/UserGuide/confused-deputy.html) documentation covers how third-party roles should be limited (read 16 Nov 2026). **What is an exposed network service?** A reachable service or request path that gave a caller a way into a workload, such as an open management port or a forged request to a metadata endpoint. AWS's [security blog](https://aws.amazon.com/blogs/security/defense-in-depth-open-firewalls-reverse-proxies-ssrf-vulnerabilities-ec2-instance-metadata-service/) describes metadata service enhancements against SSRF (read 16 Nov 2026). **What is retained access?** Access that a former employee, contractor or application kept after the reason for it ended. The ledger uses this path only when a primary source, such as a court document, says that a person kept access. It names no individual. **Does the ledger include vulnerabilities in cloud workloads?** Only when a primary source ties an incident to cloud data or a cloud tenant. A vulnerability alone is not an incident, and scanning for them is outside this ledger. **Where do the research techniques come from?** From vendors' and researchers' own publications, each logged with the publisher's name, the publication date and the link. A technique entry describes disclosed research. It is not an incident and it is not counted in the ledger. **What does the ledger say about Act's ReplicaHeist research?** It is logged as one technique entry like any other: published 3 November 2026 by act.security research, read 16 November 2026, with the link to the publication. The ledger summarizes it in two neutral sentences and does not rate it. **Are the quoted statistics the ledger's own?** No. They are third-party figures such as those in IBM's [Cost of a Data Breach](https://www.ibm.com/reports/data-breach) reports, shown as stated with year and link. The ledger does not recompute or adjust them. **Why do third-party statistics and ledger counts differ?** Surveys count what respondents report or what a vendor's customers experienced. The ledger counts only incidents with a primary source. A report may include thousands of cases the ledger cannot source, so the two measure different things. **Does a regulator action prove how an incident happened?** Only to the extent the regulator's document says so. The ledger quotes the filing for what it states, for example a consent order's findings, and records the rest as not public. A penalty amount says nothing about the access path. **How is a company's 8-K treated?** As a company statement. A Form 8-K under Item 1.05 describes a material cybersecurity incident as the company understood it on the filing date. Later amendments may change facts, so the ledger reads the amendments and shows each source's date. **Can a company ask for an entry to be removed?** An entry stays while its primary sources stay public. A company can send a correction with a primary source to editor@cloudsecuritybreaches.com. The ledger changes a fact when a primary source supports the change, and it does not remove a sourced incident on request. **Can I cite the ledger?** Yes. Use the line on the methodology page: Cloud Security Breaches, the Cloud Breach Ledger, edition 16 Nov 2026, https://cloudsecuritybreaches.com/methodology/. Cite the primary source as well when you rely on one entry. ## Related pages - [The Cloud Breach Ledger: the data, downloadable](/datasets/) - [About](/about/) - [Editorial policy](/editorial-policy/) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/faq/ --- # The Cloud Breach Ledger: public cloud incidents by access path, edition 16 Nov 2026 Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. The Cloud Breach Ledger lists public cloud incidents from 2019 to 16 November 2026. Each entry states what was exposed, how access was gained and which of eight access paths applies, using only company statements, regulator filings, court documents and provider post-mortems. A field without such a source reads not public. Every file below is this site's own data, edition 16 Nov 2026, under CC BY 4.0. The [methodology](/methodology/) explains how it was made. ## Downloads - [The Cloud Breach Ledger (CSV)](/datasets/cloud-breach-ledger-2026-11-16.csv): 111 rows, edition 16 Nov 2026 - [The Cloud Breach Ledger (JSON)](/datasets/cloud-breach-ledger-2026-11-16.json): Public cloud incidents with disclosure date, cloud, what was exposed, how access was gained, the access path, the kind of control and the primary sources. ## Cite this **Cite this.** Cite this line: 'Cloud Security Breaches, the Cloud Breach Ledger, edition 16 Nov 2026, https://cloudsecuritybreaches.com/methodology/' ## Changelog - 16 Nov 2026: first edition. Every row read from the source linked in the row on 16 Nov 2026. ## Related pages - [About](/about/) - [Editorial policy](/editorial-policy/) - [Corrections](/corrections/) ## Sources - Veradigm Inc.: https://www.sec.gov/Archives/edgar/data/1124804/000119312526385249/mdrx-20260908.htm (read 16 Nov 2026) - IEH Corporation: https://www.sec.gov/Archives/edgar/data/50292/000117494726000761/form8k-36187_iehc.htm (read 16 Nov 2026) - Amgen: https://www.sec.gov/Archives/edgar/data/318154/000031815426000119/amgn-20260729.htm (read 16 Nov 2026) - AdaptHealth Corp.: https://www.sec.gov/Archives/edgar/data/1725255/000110465926080297/ahco-20260627x8k.htm (read 16 Nov 2026) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/datasets/ --- # A sourced ledger of public cloud incidents, read by the access path that was used Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Cloud Security Breaches is a ledger of public cloud incidents from 2019 to 16 November 2026. It reads each by the access path used, lists eight access paths with provider documentation, logs disclosed research techniques, quotes third-party statistics with their sources, and holds guides built only from company statements, filings and post-mortems. The publication has one subject: how access to cloud data was gained in incidents that the affected companies, regulators, courts or providers have described in public. Each entry gives the organization, the dates, what was exposed, the access path and the kind of preventive control that addresses that path, with the source links. The method is fixed in advance. Sources are limited to primary documents, a missing fact is written as not public, and no individual is named. A statement a company makes about its own incident is labeled as the company's. The read date for the first edition is 16 Nov 2026, and each page shows when it was last reviewed. The site ranks nobody. It does not score companies or providers, and it does not rate how serious an incident was. Posture scanning, workload vulnerabilities and product rankings are outside its scope, and a page that touches them says so in one sentence. ## Related pages - [Editorial policy](/editorial-policy/) - [Corrections](/corrections/) - [Contact](/contact/) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/about/ --- # How Cloud Security Breaches is written Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Cloud Security Breaches publishes entries and guides built only from primary sources: company statements, regulator filings and decisions, court documents and provider post-mortems. Each page answers its question in a short summary, states numbers with their source and date, and ends with a source list showing the day each source was read. Entries use neutral wording: the company stated, the filing says, the post-mortem says. No entry speculates about cause, blame or motive, and none names an individual. A fact with no primary source reads not public, and a feature or detail that was looked for and not found is written as none found, with the date. Headlines state the answer in plain words. Each entry carries one verbatim quotation of at most 240 characters from its primary source, which must appear in the saved copy of that source. Research techniques are credited to the publishing organization, never to a person. Pages are reviewed when a cited source changes, when a reader reports an error, and at least once every six months. A material change is recorded on the corrections page with its date. The glossary and the FAQ follow the same rules as the entries. Definitions stand alone in one sentence, and each is tied to a source that was read and dated. ## Related pages - [Corrections](/corrections/) - [Contact](/contact/) - [Privacy](/privacy/) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/editorial-policy/ --- # Corrections and the correction rule Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Every entry has a corrections link to editor@cloudsecuritybreaches.com with the entry name in the subject. Send the primary source that shows the right fact: a company statement, regulator filing, court document or provider post-mortem. A confirmed error is fixed on the page, the review date changes and a dated note is added. Corrections are checked against the cited source and against any newer statement by the same company or regulator. If a company amends a filing or a provider updates a post-mortem after an entry was read, the entry is updated and the new read date shown. Requests that rest only on news reports or on unsourced claims cannot change an entry. ## Related pages - [Contact](/contact/) - [Privacy](/privacy/) - [Terms](/terms/) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/corrections/ --- # Contact Cloud Security Breaches Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Write to editor@cloudsecuritybreaches.com with corrections, questions about an entry, or requests to cite or reuse the ledger. For a correction, put the entry name in the subject and include the sentence you think is wrong and the primary source that shows the right fact. Messages are read by the editor. The site has no accounts, no comment sections and no contact form. Email is the only channel, which keeps the record of each correction in one place. If a message needs a reply, it is answered from the same address, and a resulting change is logged with its date. ## Related pages - [Privacy](/privacy/) - [Terms](/terms/) - [How an incident gets into the ledger](/methodology/) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/contact/ --- # Privacy Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. Cloud Security Breaches sets no cookies and runs no analytics script. The web server keeps ordinary access logs, and nothing else is collected about visitors. There are no accounts, no forms and no advertising, so the site holds no profile of any reader and sells no data to anyone. The access logs hold the address, the page requested, the time and the browser string, and are used to operate the site and to find faults. Email sent to the editor address is kept to answer it and to record corrections. There is nothing to request or delete beyond an email you sent. ## Related pages - [Terms](/terms/) - [How an incident gets into the ledger](/methodology/) - [Cloud security statistics: ledger counts and quoted figures](/statistics/) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/privacy/ --- # Terms of use Facts read 16 Nov 2026. Page reviewed 16 Nov 2026. The content of Cloud Security Breaches is provided for general information about public cloud incidents. It is not legal, audit or professional advice, and each entry reports what its primary sources stated on the date shown. Check the sources before relying on an entry for a decision or a filing. You may quote short passages with a link to the page and its date. The ledger may be cited using the line on its methodology page, and the primary source should be cited as well when you rely on a single entry. Company and provider names belong to their owners and are used to identify the sources read. Quotations from primary sources are short and attributed. The site is provided as is, and its pages may change as the sources they cite are amended. ## Related pages - [How an incident gets into the ledger](/methodology/) - [Cloud security statistics: ledger counts and quoted figures](/statistics/) - [Correct an entry in the Cloud Breach Ledger](/ledger-corrections/) Last reviewed 16 Nov 2026. Canonical: https://cloudsecuritybreaches.com/terms/